Advanced Project Management

ADVANCED PROJECT MANAGEMENT
Advanced
Project
Management
Second edition
F L HARRISON
Gower
All rights reserved. No part of this publication may be

reproduced, stored in a retrieval system, or transmitted in
any form or by any means, electronic, mechanical, photocopying, recording, orothenvise without the prior permission
of Gower Publishing Company Limited.
First published 1981
Second edition 1985
Reprinted 1986, 1987
Published by
Gower Publishing Company Limited,
Gower House, Croft Road, Aldershot,
Hants GU-I1 3HR. England
British Library Cataloguing in Publication Data

Hamson, F.L.
Advanced project management - 2nd ed.
1. Industrial project management
I. Title
HD69.P75
658.4'04
ISBN 0-566-02475-6
Typeset in Great Britain by

Guildford Graphics Limited Guildford Surrey.
Printed and bound by Billing & Sons Limited, Worcester.
Contents
List of Illustrations
Preface to Second Edition
1 Introduction
Why project management? - Traditional management theory and project management - Project
organisation - In-company project organisation Global project organisation
2
ix
...
xlll
Introduction to Planning
Managerial philosophy of planning - Functions
of planning - Level of detail - Planning the
planning process
20
3 Introduction to Control
The performance monitoring and motivational
function of control - The managerial function of
control - Managerial philosophy of control Responsibility for control - Why control of projects
is often ineffective - Conclusions
55
Estimating the Project Cost

79
Importance of cost estimating - Escalation of
costs - Inflation of project costs - The characteristic flow of estimating information - Evolution of
a project cost estima; - Contingency versus job
schedule - Cost consciousness in design - Form of
contract
CONTENTS
page
5
Planning the Smaller Project and the Basic Tools 116

The basic steps of planning - The science of
project planning - Manpower planning Budgeting - Cash flow forecasting - Bar charts CPM1PERT:network analysis - Bar charts versus
CPMIPERT - Modem computer-based techniques - Modem project planning systems Modern project planning methods - S curves Milestone charts - Management of float - Operational plan and schedule - Planning library
modules - The line of balance technique
The Planning and Control of the Larger Project, or

198
Portfolio of Smaller Projects
The major factors leading to problems
The Elements of Advanced Project Management 203

Work breakdown structure - Managing the larger
project as a collection of smaller projects Hierarchy of plans and the rolling wave concept Multi-project planning and control - Planning
the very large project
Performance Analysis
241
Control centres - Methods of analysis - Forecasting the final cost - Data overload - Data
analysis for information - Coding - Cost account
report - Higher-level reports - Traceability Control of changes to a project
Project Management Computer-based Systems

289
Essential minimum system - Types of system Conclusions
10
Planning and Control of Procurement

302
Project liaison - Buying- Inspection- Progressing
and expediting - Analysis and reports
CONTENTS
Page
11
Human Behaviour in the Project Setting

319
The authority problem in project management Authority, organisation and managerial philosophy - The engineer as a manager- Interpersonal
behaviour in the project setting - Groups in the
project setting - Team development - Intergroup
conflict in the project setting
Appendix: Planning Techniques
358
Bibliography
368
Index
371
Illustrations
Figures
Page
1.1
1.2
Matrix organisation
Global project organisation
2.1
2.2
2.3
2.4
Contribution of project planning

Control cycle
Typical petrochemical project design plan
S chart for design
21
32
49
50
3.1
Variance analysis
66
4.1
4.2
85
4.7
Effect of inflation on project expenditure

Level of uncertainty of project estimate versus
time
S chart for design
Contingency versus schedule
Distribution of project cost between phases
Relative influence of each phase on basic cost
of the project
Emphasis on cost control.
99
105
5.1
5.2
5.3
5.4
Multi-stage explosion of project

Manpower plan for critical trades
Project manpower
Productivity with overtime
119
124
125
127
4.3
4.4
4.5
4.6
89
90
95
97
ILLUSTRATIONS
Project plan for simple example

Project budget
Hysteresis cash flow
Bar chart
Arrow diagram
Time analysis
Data base for an activity
Computer-produced bar chart
Manpower chart
Plan for electrical instrument design
Time analysis for electrical design and total
drawing office
S chart for construction
S curve for electrical design
S curve for project
Milestone report week 48
Milestone chart
Skeleton bar chart showing free float
Free float monitor
Electrical designers required after
rescheduling
Drawing office plan
Draughtsmen required after rescheduling
Macroplanning module
Standard microplanning modules (A)
Standard microplanning modules (B)
WBS for a multi-unit project
Programme for a single unit in LOB example
S chart for programme
Line of balance
LOB at week 40
page
129
130
135
137
138
142
155
158
159
160
162
163
164
168
172
173
175
177
180
181
182
186
187
188
190
193
194
196
197
ILLUSTRATIONS
Arrow diagram
Work breakdown structure
WBS for a ship system
Integration of work breakdown structure and
organisation
Explosion of cost account into work and
planning packages
Matrix of responsibilities
WBS and estimates for simple project
Project plan for simple example
Project budget
A hierarchy of network plans
A hierarchy of bar charts
Cost account plan
Hierarchy of plans/cost account version
Manual calculation of project budget for
cost accounts
Calculation of budgeted cost af work
performance (BCWP)
Typical project S curve
S curve for cost account 3
Performance report for period 2
Performance report for period 3
Integration of cost and schedule
Flow chart for computer programs
Coding of WBS, OBS and cost accounts
Cost account report
Hierarchy of reports
Traceability of reports
Change request form
Integration of changes into the control and
information system
page
206
207
208
210
212
213
215
216
218
223
224
228
229
ILLUSTRATIONS
Page
9.1
9.2
A1
A2
A3
Integrated project management information

system
Microplanner program structure
293
296
Planning data base

358-9
Definitions in critical path method/nehvork
analysis
362
Dummies
364
Preface to Second Edition
Project management is a highly professional branch of

managementwhich is used in all areas of industry, commerce
and government. It involves the application of many specialist concepts and techniques but all too often peopleworking
on projects receive little or no education or training in
project management. As a result, the same mistakes are
made over and over again, and projects cost more, and take
longer, than necessary.
This book aims to bridge the gap between introductory
books on project management, planning and control, and
advanced professional practice. There are light-years of
difference between, say, a training in simple critical path
analysis adequate to plan and control a small project, and
the training and experience required to handle the problems
encountered in the management, planning and control of
the typical manufacturing or construction project. Thus, this
book is intended to be:
a professional level guide to the management, planning

and control for engineers, accountants, managers,
architects, chartered surveyors and others involved in
project work.
an advanced textbook for senior undergraduates and
postgraduate students taking the courses in project and
construction management which have been, or are
being, introduced in many universities, business schools
and colleges.
Many people drift into managing or co-ordinating projects

without any formal training, or being aware of the principles
PREFACE TO SECOND EDITION
or problems involved. They often perform satisfactorily on

the-simple, smaller projects-butas the projects become larger
or more complex they start to experience difficulties in
planning a n d controlling these projects, and to become
involved in conflict with other managers.
This is not their fault and is primarily due to their lack of
education and training in project management, and to the
fact that managing projects is very different from managing
operations. A project has to be launched, that is uniquely
organised, planned, resourced and budgeted, whereas
managing operations essentially consists of keeping an
existing organisation going. Similarly a project is never in a
steady state situation, unlike operations in theory, if not in
practice, and is always in a constant state of change.
Whereas control of operations is concerned with minor
variations about a relatively constant level of work, control
in project work is often concerned with major variations
about workwhich itself is constantly changing. Thus control
of project work is much more difficult than control of
operations.
Similarly the informal organisational forms necessary for
project work, the necessity for the project staff to cross
departmental boundary lines, the indefinite and often
informal authority of the project staff and the necessity for
the project manager to'hustle' all those involved, can all lead
to conflict Unless the project manager is aware of all these
factors and more, and has the managerial skills to deal with
people in the project setting, conflict is often unavoidable. It
is not the 'amateur' project manager's fault, it is simply
characteristic of the project environment, and can only be
avoided, or perhaps only managed, by professional project
management.
Over the last few years, our knowledge and experience of
project management, planning and control have increased
considerably, and methods have been developed to overcome
many of the problems encountered. This book describes
how these modern concepts and techniques can save time
and money. It draws on many years' experience of project
management, planning and control, developments on North
Sea oil-related projects and the US aerospace/weapons
industries, experience with modern computer-based

information systems, and organisation and human behaviour
theories.
In the context of this book, project management is defined
as:
1 Planning
2 Controlling
3 Managing people.
Emphasis is given to the following concepts, systems and
techniques:
Redefining planning
The definition of project planning is extended to cover all
those activities necessary to 'launch' a project, namely
Planning - the organisation
- the work
- the manpower
- the money
- the control and information systems.
Integration of sqstems
Emphasis is given to the necessity of the integration of all the
project subsystems, including those listed above, into one
integrated project management information system on one
structure with common cost, information or management
centres.
Control
Emphasis is given to the problems of controlling a project,
on the basis that launch planning is the prime function of
project management for only 20 per cent of the project life
cycle, with control being the prime function for the remainder.
The management, planningand control of the largerproject, or a
portfolio of smaller projects: advanced project management
The management, planning and control of the larger project
is very different from that of the individual smaller project
and another dimension has to be added to project management to cope with the problems involved.
The same problems that apply to the management of the
larger project apply to the management of a portfolio of

smaller projects. The project manager who is handling a
number of smaller projects, is actually handling a single
larger project and he should be familiarwith the concepts of
advanced project management and apply them to his total
portfolio of projects.
The other dimension involves:
Work breakdown structure The use of a formal structure to
break the project down logically and systematically into its
component parts to:
1 Enable the planning to be carried out effectively.
2 Enable the organisation structure and the work plan
to be integrated.
3 Design and integrate the control and information
system.
Management of theproject The treatment of the larger project
as a collection of smaller projects to simplify the management planning and control of large numbers of activities.
Hierarchical and rolling waveplanning The use of a hierarchical
structure of plans and a rolling wave approach to planning
the larger project, which can be extended to multiple
projects and the very large or. macro project.
Perfmance analysis A performance analysis approach to
data which produces simple to understand, but more
meaningful information for management and control.
Change control systems An emphasis on the systematic
control of change to actually manage it, whilst accepting it as
a natural part of the project management process.
The book falls naturally into three segments: an introductory segment, an advanced segment and a people segment
The introductory segment deals with the general principles
and judgemental facets of project planning and control, and
introduces the basic planning techniques in the context of
the smaller project. The book begins by describing the
different forms of organisation structure used in project
work, and enlarges on the'matrix' form for single and multi-
PREFACE TO SECOND EDlTION
company project organisations. The following chapters deal

with how to organise the planning process. They discuss
who should do the planning, the role of the planner in 'realtime' planning, and the common misapplications of planning
techniques, and describe the characteristic evolutionary lifecycle of project planning. An entirely new chapter on the
control of projects is introduced, which discusses the functions and responsibilities for control; why control is often
ineffective and outlines the requirements and standards of a
good project control and information system. This part of
the book also looks at the importance of manpower planning,
critically reviews and compares the basic planning techniques and outlines the difficulties involved in applying
these techniques to the larger project. It outlines the requirements of modern project computing systems and planning
packages and describes how these systems can be used in
conjunction with such techniques as milestone charting and
S charts. The use of planning library modules, multi-project
planning and control, and the line of balance technique are
all explored in detail. It includes a review of estimating the
importance of cost consciousness in design and the normal
forms of contract used.
The advanced segment deals with the more sophisticated
modem concepts of project planning and control, as outlined
above, in the context of the larger project.
The people segment, the final chapter, examines the all
important problems of human relations in the context of
project management, including the basis for the 'authority'
of the project manager, methods of coping with 'we-they'
attitudes and conflict, and how to develop teamwork.
Acknowledgements
Many people have contributed, directly or indirectly to this

book, including my secretary, Elsie, who has deciphered my
scrawl to produce the manuscript. In particular I should like
to thank my wife, Margaret, who has put up with my
xvii
disappearing for long hours in the evening or at the

weekends to work on 'my book', but who through it all has
encouraged me to persevere with it.
F. L. Harrison
xviii
1 Introduction
--
A project can be defined as a non-routine, non-repetitive,

'one-oft undertaking, normally with discrete time,-financial
and-technical perfopiiance goals. Since the Iate 1950s the
terms project engineer, project co-ordinator, programme or
project manager have become common and are signs of the
increasing use of projects in all areas of industry, commerce
and government.
Organisations using projects are, knowingly or unknowingly, involved in a relatively new concept of management,
namely, project management The management of projects
is very different from the normal management of operations,
and involves specialised forms of organisation and methods
of planning and control. Training and experience in general
management, or even in construction management, are not
sufficient for effective project management, which has
grown to become a separate, highly sophisticated area of
management. The definition of the project management
concept as a specialist area of management came about
initially in the US Aerospace Weapons Research and
Development Industries. The Manhattan Project during
World War 11 was an early example, but it was not until the
1950s that project management began to crystallise into its
present form. The problems oftaking an aircraft, orweapons
system from early research through technical development,
design and manufacture are fraught with complexity,
uncertainties and difficulties. This is particularly so when
working on the frontiers of technology. Planning and cost
control in these kinds of projects are very difficult, and
different organisation structures, authority and responsibility
patterns, and specialist techniques are necessary. Other

important users of the project management concept, from its
earliest days, are the oil and chemical industries, with the
engineering contractors involved in these industries. The
management of engineering, design and construction projects in these process industries involves many of the same
difficulties as in aerospace weapons projects.
These two industries are todav urobablv
the most vro*
fessional in the application of the project management
conceot. and make widesuread use of the most sovhisticated
te~hn;~ues.
~rofessionaf~roject
managers in t6e large oil
and chemical design and construction contractors are
employed on an international, worldwide basis.
Large civil engineering contractors have also implicitly
employed the project manager concept in contract management for centuries. However. in the vast their uroiects
tended to be not as complex as the aero&ace/weapons and
oiVchemica1 industry projects, and specialised organisation
structures, planning and control techniques were not
explicitly recognised or used. Today this has changed, and
project management concepts and techniques are in common
use in the civil engineering industry.
The project concept is also used in high technology
research and development, in the engineering and manufacturing industries for new product development and
manufacture, and in the design and construction of new
plant and equipment. Building a new steel plant, manufacturing a boiler for a power station, adding a plant
extension, developing and beginning the manufacture of a
new product, designing, manufacturing and installing a
heating and ventilation system are all projects with similar
problems and require similar management and techniques.
The design and introduction of computer systems, and the
development and introduction of new suites of computer
program also uses the project management concept. Thus
the computing industry and the computing function within
companies, both employ project management
Outside the engineering field we find that the concept of
product management in marketing is very similar to that of
project management; the organisational forms, managerial
<
INTRODUCTION
problems, planning and control techniques are almost

identical. Similarlv, the introduction of new products, the
feasibility study fdr a company takeover, andthe irnprovement of industrial relations in a comvany can all be treated
as projects, and can benefit from ihe ;se of the project
management concepts and techniques.
Many firms who have traditionally not been involved in
project work are being forced to adopt project management
concepts. This is partly because of the behefits obtained, but
principally because customers, particularly foreign customers, are insisting on one firm taking overall responsibility
for all the work involved in manufacturing, supplying and,
or, constructing large and complex orders or contracts.
Therefore vroiects.
and as a conseauence (exvlicitly
or
"
. implicitly) project management, are befng increasingly used
in all areas of endneering, construction, manufacturing,
research and devilopmeG. These projects have similar
management needs and problems, and thus the concepts
and techniques of project management transcend industry
barriers and all areas of engineering.
Projects and project management are used in some form
or other, on many small scale ventures with good results, but it
is on large scale applications that good project management
is a really significant factor. he problems of man-manage__
p anning and control,,where
ment, financial-management,
- _
many
firms
are
involved,
wifiToiig
time spans, large
- unartainties-Cnd very large sums of money, can lead to
inefficiency~eTaiXiFc~mpletion
and wasted money, unless
eFEctlxproject management
is
used.
-
5-
___-
--
----
Why project management?

The principal reason for the development of the project
managemekt concept, organisation, and specialised, often
highly sophisticated techniques, is that the traditional forms
oforganisation structure and management techniques do
not handle project type work effectively. There is a need for
different forms of organisation, specialised information
systems, managers skilled in the techniques of project
planning financial management control and the particular

human problem arising in project work because of the
special characteristics of projects and the problems caused
by them.
P~ojectsare essentially temporary-activities-for those
concerned, withtypical lives of from six months to five years.
hJanagemenLorganisation-and information systems have to
be established anew for each project, and there can also be
o n 5 a veryfimited learning curve for-those involved.
~ < Z a ~ manager
sa
may only pass through each phaseof a
project once, every year to five years. In addition, decision
making in project work tends not to be repetitive and bad
decisions at any stage in a project affect it throughout its life,
and it is not generally possible to--recover-f-early
d e f i c ~ n c i e ~ ~ ~ ~ r o j ~ ~ n ~ g ~ Normally, projects-involve several departments of a
company working together and in the majority of cases more
than one company is involved in work on one project. Often
these departments, or companies are working on several
projects at the same time, each at different stages in the
project life cycle. Project work is therefore necessarily
complex with respect to interdependencies between the
departments and companies involved. This involves, explicitly, a complex organisation, including people from many
different professions, backgrounds, trades, departments and
companies. Additionally, these relationships and interdependencies are dynamic and never static. Typically at the
start of a ~roiect,
<
. work emuhasis may be in research and
development; it then changes to engineering design, to
uurchasing and urocurement, to manufacturing and, or
constructi&, to testing and commissioning, and-finally to
operation. No one functional department or company is the
most important over the whole life of the project and thus no
individual functional manager can assume the leading
management role for the complete project.
Because projects are unique one-off undertakings, there
are problems in defining work, organisation, allocating
responsibility, budgeting, planning, control, communication
and co-ordination. M a x p~~~c~s-haye-sufferepdiffmlties
because of a la& of clear de ____.inition of orgam~onstructure,--
-__
INTRODUCTION
which in turn compounds these problems. Great em~ha@s

must be given in-project-work~o~heplanning
and control
acTiCifiei, and manyprojects-have-sufferedsIEl@s-and over--
exp~nditu~6e~use~of~inadequ~te~plan-g~and~controlsystems. Shecia1 project-oriented information systems are

also required for effective project communication, coordination and control, and these systems are different from,
and generally have to be separate from, the information
systems in the normal traditional management organisation.
There are also particular difficulties involved in achieving
effective communication and co-ordination with several
different organisations involved in a project, and as projects
get larger and more complex, communication and coordination become more and more difficult, and yet more
vital to the success of project work
Usually, project work involves large capita1 expenditures,
and the financial management and control of projects is
thus extremely important, to minimise the overall cost of the
project.
Finally, the temporary, complex and often loose nature of
the relationships and authority patterns involved in project
work, combined with the number of different departments
and companies involved in any one project, whose objectives
and management styles may differ, leads to human behaviour
problems and a tendency for conflict between groups and
individuals. Thus the traditional management theo~yand
organisational structures have to be modified in project
management.
Traditional management theory and project management

The
traditinnal-m~del -of a n - o r g a n i s a t i o n a m e i s
-based upon t h r ~ o r g a n i s a t i o n a l c o n c e p t s ~ w h i c ~ l l
vXat7diii
E e management
of projects,
namely
---1 The functional division of management.
2 The hierarchical concept of superior - subordinate
relationships.
3 A number of so-called principles of management.
The conventional_fo~m
ofcompany>rganisation divides the
@pplein the companyjntogroups of similar skills, interest,
t r a i n i z anZ-Ecupational specialisation. Therefore one
normal$fin& a company organised into departments, such
as engineering design, construction, plant operations,
purchasing, marketing and finance. Such an or anisation
u&injroject - work
use of la%%ur on
- -permits-the-effic~en
-- several projects, as-more or less people can be-allocatedfo
projects as required It also increases the rate of development
of professio~lalskills by constant contact and interactionwith colleagues in the same specialism. It permits the
division of labour by specialism; for example, it is possible to
have electronics, electrical and instrument engineers in the
same department They can then be allocatedto each project
in turn as required, but retain their departmental base. Thus
the functional organisation permits the efficient use and
development of resources on multi-project work
The-functional organisation thus gives the organisation a
h p h l - d i m e n s i o n , into which it is divided by departments, and added to this there is a vertical dimension, with
dferent levels, which represent varying degrees of authority.
m i e r a x h i c a l structure is the basic framework of the
organisation and-fhe Supefioi - stibordinate relationships
a?e the_lines~in~w~ich~authority~flow~from
the top managemj%tta_thethe_lowestlevels. Grafted on to this, in all B u t t k smallest company, F a division of organisational elements
into line and staff groups. Line managers have a direct
responsibility for achieving the goals of the organisation,
decision making and exercising authority. Staff members
advise and counsel the line managers and have no legal
authority, except within their own functional department,
and are typically professional specialists. This structure
forms the traditional management vvramid with the onlv
focal point of power bindingthe org&isationtogether being
the tov management of the comvanv, or division. _This
- 7 -
~~.satien-i~bests~dtdheh31.nd~ing~ffaa~ont
flow of basically repetitive-work-with each-department
w o r E n r o n its own futiction. This is the situation in the
&ical fact=-ipiTrtment
or office and leads to stable
interpersonnel and departmental relationships.
Unfortunately
this traditional form of organisation does
--\-----c---
INTRODUCTION
ot handle projects efficiently, tends not to meet schedules

&-p e r a t i o n s ~ o r i O a t ~ ~ n d n d n o t ~orogoal-oriented.
ject
There are severe difficulties involved in achieving effective
communication, collaboration, co-ordination and control
with several different functional departments involved in a
project with this form of organisation. The differences in
activity, philosophy, education, training an?TjjE~nalVaxes
a&TGiid to-incTease the likeliho&f oFconflict and misundi?istandiigbetweenfunctions,when.engagedin project
work unde-rr3i~nd_c_ost_pxessures.
In addition this traditional form of managerial organisation just cannot handle the dynamic, everchanging relationship and the complexities involved in project work. In large
projects and in those industries where development, design,
procurement and construction or manufacture must often
overlap to achieve the shortest possible project time this is a
particular problem. Effectiyecamunication and co:
prdi-nation of effort among departments and Gmpaniesare-very ifiEEiiIf to achieve wifi--thXXandard functional6~axsatiEinsin project_work,andyet lackof thesefunctions
i s o m - G ~ ~ u t ta_failu're,especr;dKy
e s
where-several
companies, for example owner.-contractors and sub-contractors, a@ involveh in onegroject. The traditional
o?ganisa-TToEKas n o m e Z i s ~ m a n a ~ ior
n eco-ordinating:its
efforts through all stages of the w&kult has no mean: of
integrating different departments, at levels below top
management, customers, contractors, sub-contractors,
material and equipment suppliers into one organisation.
Thus a special project form of organisation is required to tie
together these separate organisations into one global
organisation.
form of organisation is-required-from
the
,A -different
t + a d i t i o n a l g _ y r a y r a ~ w ~ ~ s cthe
en ind1XuaI
d
d e ~ r t _ ~ d i ~ s i o - n ~ r _ c C oL
m
~npthe
a nextreme
y ~ -case,
kundredsof-compa_nies involved.
To back this up, a project
management information system-is required for each individual project, which is normally separate from the normal
operational management information system of a company.
This form
- --of organisation has come to be called the project,
or 'matrixl_organisatiOnT
Even with this form of project organisation, there is the
-'
--
--__
_ -__-
%
-
need for a separate management function to manage projects.

The senior general management of a company cannot give
the concentrated attention required for the effective management of any individual project. This is particularly so, when
a stream of projects is being handled, each in a different
stage of the project life cycle, for example, development,
design, manufacture or construction. With the traditional
departmental form of organisation, no one manager is
entirely responsible for, and manages a project throughout
its life. Emphasis by functional managers is often given
more to the technical success of the project, rather than to
the time or financial success. Functional departmental
managers tend to concentrate their efforts and understanding
within their own departments. This often leads to tunnel
vision, that is, they can only view a projectwithin the narrow
scope of their own function and cannot exercise judgement
on the overall project.
Thwe-is thus a need forvesting in one individual or group,
responsibiiityG~Finte~aZii~
all th<miariagSment functions
anCatiivities?~fFie many sub-organisations involved in the
prQjecLundeiTaEinglTheman responsible for a project may
be called a project manager, have complete control over all
aspects of his project and manage a large organisation. At
the other end of the scale, he may be called a project engineer
or co-ordinator, with only part of his time spent in management duties, have very little authority and no staff. Both
these people have the same responsibilities, and their
management duties cover the same critical areas of management
The project manager concept has thus evolved to be a
general management and integrationist activity, below the
level of top management. It provides the individual responsibility, management and co-ordination for any one
project, which is essential for the success of any such
undertaking. T h J a m a 4 a g e r - i s r e s p o n s i b l e for the
success of hisgroject-in termspf t i m e , c o ~ ~ c h l i I c a l
performance. He must provide_thema_nagementand leaderAsh@>-=to-bind
the peopleand S u i m T f f e r e n t
departments and companies working on the project, into
&r-ma~aagffial
organisation -and t e a m and provide-the
--
INTRODUCTION
--
drive necessary to ensure completion on time and within

-
=us
the project management concept endeavours to
provide at the minimum, the co-ordination of the workof the
different groups involved and, at the maximum, the integration of all the groups involved into one effective organisation.
The characteristics of project management, which also
cause it to differ from conventional management theory and
which clash with the conventional principles of management
theory, are that the project manager must manage people in
his own company functional departments, and outside
companies, from a position sometimes junior to department
heads, and normally outside the traditional pyramid structure of authority. He deals vertical1 , horizontally
-and
*n3
w i t h p e e ~ ~ ~ & l . a stheicsupEiors.and
- subordinates. No matter how necessary the horizontal
a n h d i a g o n z acEities are, relationships are difficult to
formally define, and the project manager usually has
indefinite, or little direct authority over all the people
working on the project.
Because of the distinctive characteristics of projects and
work on them, these project personnel must also have
specialised skills and knowledge of the techniques used in
project planning, financial management, and control. In
addition, they must have skills in handling the human
problems which arise because of the characteristics of
projects and also because of the particular problems associated with the project organisation. In this, several of the socaled-psinciples,of managemeen-<-such as-aur' .
USL
egua&sponsibilityty~d a subordinate must o n m o n e
superior, m$t 'be disregarded, with consequential probIems
-,
1.n-humCZ%-lations.
The project manager has to have some knowledge of, and
judgement of, the functions and disciplines contributing to a
project, but above all he has to be a skilled manager. Unfortunately, the education and training of engineers is generally
confined to technical functions and does not normallv
encompass organisation structure. !nanagement information svstems. cnrnmunication and co-GGnaTioi;.. ssciaIised
.
-ng,
rinancial management and control techniques,
t
10
and the human relation skills required. These functions are

essential to the success of project work, and it is in these
areas a project manager must be a specialist.
Thus the management of projects is a difficult form of
man&ement and many projects have suffered in the past
from inefficient project management. resulting in increased
cost, delays in completion and managerial conflict. Recent
Sea oilexperience, with particular emphasis
. on
. Norths in
which
related projects, has shown that the
shortcomings in p r o j e m a n a e t e to cost
-e=alati~nand-delays in the completion oErojects
_--are as
_._
follows:
-_
_-
1 The general management and logistics problems

arising from the complex interactive nature of project
work
2 Lack of clear definition of organisation structure,
which causes problems of authority, responsibility,
communication and co-ordination.
3 Inadequate planning, budgeting and control systems,
not inteerated with the management structure.
4 The intergroup, interpersonal problems and conflicts
that arise in the flexible and complex organisation of
projects.
However, effective project management has been successful
in handling the many planning, communication and coordination problems involved in project work It has proved
capable of integrating the various personnel and groups into
one organisation and, due to the nature of project work,
developing effective team work and commitment to the
project objectives. It is now widely accepted, implicitly or
explicitly, that some form of project management is necessary
where large and complex undertakings are involved, and
that its effectiveness can significantly influence the cost of a
project and the time taken to complete a project.
v
Project organisation
The form of organisation used to manage projects is often

indefinite, or informal. Yet recognition of whaLorganisa-
11
INTRODUCnON
organisation structure be related to the work to be done on

the project, and that it be integrated with the planning and
control systems.
Many projects are large and complex, and involve several
contributing organisations, some of which are normally
separate companies. In such pr~ject~~the_organi.sation
structure must,
.
in. turn
- . a--m i t X
. . l y be comp_lex,a_ndjt-is
us'e%mook_at_tke-structure
... -- .__ of the
__ project-organisation-on
_-.
twoscales:
namely,She~ro~ect
organisation-within
- ~.
a single
. .~ ~.-.--.
that is, in-companygroject_organisation,
-and the
company,
organisation
into
global project
.-.-. .----&ing several organis_at@ns
one project organlsaio,nl
-.
1
-
~-
--
-.
=
~.
_I-<
In-company project organisation

There are three_f ~ x m s _ o f p r o ~ eorganisation
ct
in use today,
~ ~ F ~ i d ~basic-difference
l ~ i n gbetween them-being the
5 o - ~ n i aufi5iity
f
that is given to t h s e r s o n responsible
for
the
project,-nim*
-
__
I
_
1 A line and staff organisation, i.e. a project coordinator.

2 A divisional organisation.
3 The 'matrix' organisation.
Project co-ordinator
The first stage of evolution to some form of project-oriented

organisation is to appoint someone, whose prime responsibility is the project, to co-ordinate the work on the project of
the people in the functional departments. In this form of
project organisation,
-- - - the-person responsible% the proJ'ect
ags&&ntially in a staff position to support line managementz-thZ fTi~1cticma1
department, and retains the tradi- -
->
12
tional line and staff functional organisation essentially

uncuag~-is-form
of project organisation exists in three
variations, where the person responsible:
1 Is a staff assistant to a senior general manager, and

co-ordinates work on the project.
2 Is a project engineer, for example, in a factory,
responsible for works projects, carried out by incompany personnel and sub-contractors.
3 Is a functional member of a group of people working
on aproject, and is given the additional responsibility
of co-ordinating the work of the group.
The emphasis in this form of project organisation structure
is on co-ordination, communication, planning and control.
The pro'ect co-ordinator acts as a focal point for information
o d p ? and-l"sl"ss'eZn
~ ~ akin-an
~ ~ advjsoystaff pGtion,
w-eTG-3
ihE general manager, but not having line
,_management authority and direct responsibility for the
project.
This position generally has little status, and is normally
o ~ & i ~ Z ~ ~ ~ < g T junior
a t i v eengineer,
I~
w5h litile authority
or experience and training in specialist project management,
planning and control techniques. T&sthe projectis-not
managed, it is co-ordinated. The project co-_ordinator-can
onl~~moniGor
progress, practice a%ZE&
of co-ordination,
givxthe goodwill of thefunctional persqnnnljnvolved, and
is in t h e - f r ~ X o ~ c $ a ~ u p p l itoa nthe
t managers of the functional departments. He cannot-effectivelgbring pressure to
bear, or exert authority_toAobtainaction. He must persuade
others, from a position of weakness, and feels very insecure
and frustrated, and is generally unsure of his role.
He finds it very difficult to manage, or exert leadership on
the project, and control is normally weak When evidence of
an early trend is slight and inconclusive, he cannot normally
get action. I t i ~ o n l ywhen adverse trends are patently
obvious, that he can get chiiige implemented,=d-this is
normaIw tool_lf&e.Thus; _this form of organisation can
achieveco-ordination
- of pr6ject work, but loses many of
the a_dxanta@of the project m a n a e m e ~ m n c e p k i t n dis
----only--suitable
for small projects-normally
- \ I
INTRODUCTION
13
Divisional project organisation

At the other end of the spectrum is the divisianalpmject
6 r- --S ---n W i i-ise 3rongZt f o r z o f project organisation,Personnel are allocated completely to t h ~ ~ r o j e c t - ~
organisafi-ife --0f-thip6ject and-the project
manager has fuIninemaiLa@ment authority,-at least over
people ff6m Eis own company. This is in effect, setting up a
separategoal-oznted division of the company, with its own
functional departments. The project manager still has the
problem of managing and co-ordinating the other companies
contributing to his project, but he is the complete master of
his own company organisation.
management
This goal-orie~ted~organisati_on_makes_the
o ~ ~ ~ o ~ e c t s + . a n d ~ t h e ~ l a n n i n g a n d . c o nWith-it
troaser
th-ere is b e t t e c i n t e d o n of those involved,comrnu~cation,
both formal and-informal ismore_frequent_a.ndthe_reis-mor_e_
c o m m i t m e ~ t o ~ t h ~ p r o j e c t ~ o b j e c t i vthus
e s . ~tends
t
to
maximise the probability of completing the project on time
and to budget.
Unfortunately it tends to use-thesc-arce-resources-of-acompany
--- - - l e-s ~ e w n i - 2 - n e c e s ~ a ~ i n - t hform
i s - of
project organisatfon~todupl~cate-the specialistsineach
project-and allocate them in total to each project, more or
less for the life of the pFoje%-Itis3lCoaifficult; o f V e i
e x m e i n f i e adibision of labour within a speciality,
and a divisional organised project may have to make do with
one electrical engineer instead of three specialists.
Therefore, though the separate divisional form of project
organisation enables projects to be managed more effectively
than the functional and project co-ordinator forms, and it
avoids some of the problems of the more complex forms of
organisation, it can only generally be used when a company
is handling a single important project, or where one project
is so very much more important than others, that it justifies
setting up a completely separate company division. It cannot
normally be used where a c o m p a _ n Y Y i s _ h a n d l ~ n g ~ ~
p G j E E T n ~ o ~ n U o i E T % s i as
s , it splinters-up-the
functioFaP and specialist -dgpartGefifs~This splintering
reduces t h F n i e r of proFcts that Company can handle,
---
14
ADVANCED PROJECT M A NA G E MENT
and it inhibits the transfer of personnel from one project to

another as required. There are good reasons for retaining the
functional departments intact, not the least of these are that
the heads of functional departments are probably long
established, exert considerable political power and do not
wish to see their departments splintered among projects. In
addition, it is an advantage to have these intact functional
departments form a pool, from which specialist support can
be drawn when required, and which permits the training
and development of functional expertise to develop much
faster in them, than if they were splintered up.
Thus the functional form of organisation is more efficient
in the use of resources, but less effective in meeting aroiect
"
objectives, whereas the divisional form of project organisation is less efficient in the use of resources, but more effective
in meeting project objectives. What is required in project
work is an organisation which is both efficient in the use of
resources and effective in meeting project objectives, and
this is what the third form of project organisation, the matrix
organisation endeavours to do, although at the expense of
organisational complexity.
- A
Matrix organisation
The organisational form, which predominates in the incompany project organisation, and is used almost exclusively
for the global project organisation, has come to be called the
.--- that lines of authority
'matrix' organisation. Thiaecognises
and responsibility, commu~nicatonand co-ordination can
exTsiofiZo3a~lyaXdddiagonally, as-wellasina traditional
vertkal_pyramid form- In this f~irn,~ersonnel_workin~
on
th_=.jectPhave i n addition to their responsibility to their
own_funCtiConal3uperiorand, or separate c o m p a n i ponsibility, or loyalty. tome proje5t and to the man m)3roup
appointedts manage tfieApr6Ject.There may be several such-groups involved in any one project, for example, for the
client or owner, contractors, sub-contractor, consultant and
equipment suppliers.
~n~hismatrix_fom.organisation,authority
is shared
between the-project manager and functional managers.
*
p d e c t manager is not the complete master of decisions

,affg&Kh&
--pGeX,IbToper~tes=^ln-a-~ s i ~ a k m ~
.&- I$YI
His a u t h o ~ ~ e ~ c ~ u ~ c r O S S ~ d E 6
vertical line and staff authority patterns and leads to both
horizontal and vertical lines of command
A typical organisation chart for a matrix project organisation is shown in Figure 1.1. It is essentially a supplementary
organisation which integrates people from the primary
functional organisation for the purposes of goal achievement. h h o ' e c t manager in the matrixxfor~ac~s~asr+junior
general m a n a ~ c 6 i K p i n work
y
is done-by the
v ' a ~ s T n c t i o f G f trepTZtrtlii-eiits,-fomample,
f
design,
purchasingaccoun~ng,constructionfor manufactuingand
o ~ r a t i o n sIt. maintains these functional departments intact
and -permits the development of functional expertise,
specialisation and the transfer of staff between projects.
Department members can be allocated to a project for its
life, orpart of its life, on a full-time or part-time basis, but still
remain part of the functional department. At the same time,
the project relationship integrates them on the other dimension, provides a means of communication and co-ordination,
management and leadership, facilitates planning and
control, and gains their involvement in the project and their
commitment to meeting the project objectives. By making
the individual a member of both a functional and goaloriented organisation it endeavours to obtain the advantages
of both.
b~ The_pr~ject~manager~m~~~rel~on~support
and services
frara_theseJunctional dep-artments. k d e t e e i n e s , in consultation
- -- with their managers, what is required and &%%en,
and the fufct&~alfaaTager cEitrroTsfiow and bywhmii.The
- functional line managers can do a3:~jobbof alto*
t&ir own departmental personnel-to the-various projects
they are working on, than the project manager, as they know
the capabilities of their own men and the resources available.
he functional manager also has the responsibility for
technical decisions within his department, and the project
manager cannofcompete_with-them in their own area of
speCEalis%ifioYi. He can bring pressure to bear regarding
schedules and ' b-x g e t q T n i r a s k E r l R - m l u a t i o n or
-=
- - - __--
INTRODUCTION
17
assessment of alternatives, when there are problems. Nevertheless, responsibility for specific technical decisions lies
with the functional line managers, as they know their own
work best, are doing it, and are responsible for it.
The matrix form of project organisation conflicts with
traditional organisation theory in many ways. It has inherent
in it, dual subordination, division of authority, responsibility
without corresponding authority and a disregard for hierarchical principles.
These basic principles of organisation, with which it
conflicts, are no more than generalisations about what has
been observed to work in practice, based on past organisational experience. Being derived from.experience, they are
subject to revision in the light of new experiences and
circumstances. It is as foolish to be totally bound by past
experience, as to ignore it.
Disregard for these principles does mean that the matrix
organisation has organisational complexity and an inherent
conflict situation, but in large projects, involving different
departments, organisations and companies, there is a new
set of circumstances. In order to bind all these diverse
elements into one organisation, committed to complete a
project to its time, cost and performance objectives, and to
effectively manage it, a new set of ground rules has had to be
evolved.
Global project organisation

There is no real alternative to a form of matrix organisation
for the global project organisation, as no other organisational concept is applicable, and can tie together all the
separate organisations involved, into one organisation.
This viewpoint, that all involved, client, contractor, consultant, architect manufacturing manager, sub-contractor,
material and equipment supplier, belongto one organisation
is critical to the success of project work If the performance.
or lack of it, of any group or company involved in the project,
can influence the overall project's success, then thc principal
project managers involveh must consider them as belonging
18
to the project organisation. Thus project management,

communication, co-ordination, planning and control must
be extended to embrace all these contributing organisations
and integrate them into one global organisation.
These organisations ar*dtogethe~byJines-o_f
authority,
or>%nce,-which
_are sometimes very weak, and often
based merely- ~ n ~ c o n t r a c t u r and-purchase-order
al
agreemGtsTThus the form of contract and conditions ofpurchase
ordersare lines ognfluence-whichdetermine how the global
organisation-operates.
This global organisation structure is therefore of necessity
camplex, often loose,aild f a l i - ~ f ' ~ I l n e s . - ~ e ~ C r g a n i
tion chart for any particular project differs, depending on
the number of principal contractors. who does what, consultants involved. and also differs on the point of view, that is,
from the client's project manager, or principal contractors'
project manager. One example is shown in Figure 1.2. This
shows the principal lines of the project relationships only,
and ignores the functional, departmental and company
lines of command.
In addition to these project hierarchical relationships,
there are normally relationships between similar functional
groups in the different sub-organisations, or company. For
example, the client's design engineers will communicate,
and perhaps supervise the contractor's design work, and the
client's construction group will similarly be involved with
the contractor's construction group.
2 Introduction to Planning
it.ies,~esources required, con~ ~ h i p s - c a n 3 e = S a m - ~ h e

wdd-dplamingcmbe-&nZinFormalmhe
head-QLO~the back ofanenvelope. Even on large projects
the various stages of the work sometimes have a simple,
logical sequence and planning can be done informally.
However, when a project gets over a low threshold in size or
complexity, it is impossible for the human brain to take it all
in, and some formal planning system is necessary to ensure
efficient working. As projects get bigger and more complex,
planning and control become more difficult but much more
essential to their efficiency and success.
work normally follows
In the management o-ions,
a w m T l & l routine; people know their job and carry out
basicallv the same work week in. week out. Relationships
and infiwathnfllows are-more or less p e r m a n ~ ~ a ~ h &
is an effective leamine curve brought about by work repetitign. Monthly budgetlare straightForward measurements of
variTbles, that change only slightly from period to period.
Operational management can thus be viewed in the short
term as a relatively static situation.
None of this applies to project work! A project is by itsvery
d e f ? n i t i ~ ~ n - ~ E i i i undertaking;
iie
thus in all but the
simplest project, the people, the companies, the work, the
management information system, and many other-factors
reeuire to-be organised uniquely in temporary but ever
ch-angingelationships.Thus in projectwork, planning, that
is deciding who does what, when, how and why, is an
essential aid to organising the companies, the departments
INTRODUCTION TO PLANNING
21
and the people involved, and to executing, managing and

controlling the work on a project Without effective planningin projectwork, there is chaos.
~ I e v ~ r ~ l a n n i n ~ s ~essential
n o t ~ 5l ythe-organisation,
decision
miikingand
resourCeTocaFioG
on a project,
-but is also required to integrate and co-ordinate thk efforts of
~ ~ ~ o p l e - w o r k i n ~ o n n a a p r o j e censure
t ; _ t ogood
n;-tpgromote time-and cGt-consciousness; and t 6
cSitir1 Ute to the autho;ity of the project manager. It is also
essential to establish the base lines for the control of the
project and to guide it throughout its life. Without effective
planning there is no base line for control, and the projectwill
inevitably take longer and cost more than would otherwise
be the case. Figure 2.1 shows the contribution of planning to
project management.
In recent years such planning has often been associated
with the use of advanced techni~ues.These techniques have
not only made planningvery fashionable, but they have also
made it ~ossibleto ~ l a nand control large
- and complex
projects effectively.
s
"
-%?
Planning is a means oE
1 Organising the work on a project
2 Deciding who does what when how and Tor how much.
3 Determining the resources required.
4 Allocating these resources on a time-phased basis.
5 Allocating and defining responsibility.
6 Communicating between all those involved on a project
7 Cc-ordinating all the activities and people involved.
8 Controlling progress.
9 Estimating time to completion.
10 Handling unexpected events and changes.
It is also a
(a) basis for the authority of the project manager.
(b) basis for the budgeting and financial control of a project
(c) basis for self analysis and learning i.e. real experience.
(d) means of orienting people to I w k ahead. and
(e) way of initiating and mainlaining a sense or urgency that is time
con~ciou~ne~s.
Figure 2.1
Contribution of project planning
22
Unfortunately the development of sophisticated techniques that have made it possible to plan and control large
and complex projects, has also led to a concentration on
these techniques and an inadequate recognition of the
human factors involved. As a result, project planning
systems often fail, many people have become disillusioned
with formal planning, and a great many firms do not use
planning as effectively as they should. In many organisations
no formal planning is done, and in others the plans
produced are not used to organise and control the work.
Advanced planning techniques are used as sales factors, or
because the head office or client insisted on them. Complicated looking plans are produced, which are not used to
manage the work on the project, and which serve only to
impress the uninitiated and decorate walls. It is most
important to realise that for the application of any form of
planning to be fully effective, the plan itself is nothing more
than information written on paper.
Unless
-- t h e ~ k e s p o n s i b l efor executing the work
operate in accordance with this infom-ation; nothingwill be
achieved by-this
-form of planning I n addition, many
managers tend to take a short-term view of planning and
basically plan only for the immediate future. They cannot
see the overall situation, problems and constraints involved
and how what they are, or are not, doing today will affect
other areas on the project, and the project itself in three
months, or three years time. Thus, unfortunately, many
unrealistic plans are produced which are not used by the key
personnel involved, but almost just as importantly, cause the
whole concept of formal planning to be held in disrespect
There are two reasons for this: one is the misapplication of
the techniques of planning, that is, the 'science' of planning;
and the other consists of several factors more concerned
with how these techniques are used and could be termed the
'art' of planning. I u h e
nd-early 1970s project
p l a n n i ~camcto b e - a s s t h the useof thhcritical
p m- m d h ~ & ( ~ ~ ~ - d_ p r o j e c ~ a l u a t i o n - a n d ~ r e v i e ~
--ltechnique
(PERT). For a number of years it was fashionable
t o X o & c e these relatively sophisticated techniques and
they contributed much to the more effective planning of
-_
23
projects. It was thought that in all but the smallest project

CPMIPERT had to be used, as no other method including
bar charts, could cope with the number of activities involved
and the complex interrelationship.
Unfortunately many people experienced difficulties with
the use of these techniques and rejected them, returning to
the use of the older and simpler bar chart method for project
planning. This was principally because the effective use of
CPMIPERT needs much more than a straightforward
implementation of the technique, but also because of a lack
of appreciation of the art of planning. Not only is training
and practice required in the essentially numerical implementation of the technique, but also in its use in conjunction
with other techniques on large projects and an appreciation
of the problems encountered in its use, as discussed in
Chapter 5.
In
--.-.
practice, in-all but the simplerproject, no one technique
b disadvantages
is completely effective onits o w ~ E a c has
anihnTtatio5s. and normally several or even all of these
common techniques are used to compliment each other.
Essentially the project manager or planner should use a
contingency approach to planning, that is the technique or
techniques which work best for those involved, and which
are appropriate to the job in hand. Just as it depends on the
size and complexity of a calculation, whether a manager
would use a calculator, spread sheet or a computer model, so
the size and complexity of the project should determine
which technique or techniques he should use to plan it.
However, there is little point in using sophisticated techniques, which are not understood by the managers and
engineers involved, and thus will be used wrongly, or not at
all. The techniques must take into account their acceptability
by the people concerned.
The decision as to which technique should be used must
be based upon knowledge, and the project manager or
project planner should have a familiarity with all the
common techniques used in project planning. All the other
Fanage!s and sup~jsorsjnvolved~m~st~also~have~some
amiliar~ty-withthe-techn~que~actually~used~to
_plan-the
T
project. Thus all levels of management from foreman to
-----.
24
managing director should have some training in the

methodology and techniques of planning used in the project
The advantages, disadvantages and limitations of the
various techniques of project planning are covered more
fully in Chapter 5, whereas those factors described as the art
of planning will be outlined in this chapter.
Knowledge and application of the techniques of planning
are not sufficient by themselves to ensure the effective
planning and management of projects. Just as importantif
not more so, irrespective of the techniques used, are*
adoption-ofamanage_rial philosophy, which accepts that
p G g is an integral @i3XfTiZfufuncfi0noofall those
c o G ~ I r i i e ~ r o j e c t ~ g a i n ~ nhnderstaniing
g~he
orthose
i-n~olved_of-howandwhen to apply the various planning
and cannot achieve; and
techniques and just what th-n
how t o Tu3d_andduse-aglan~o-manage-a
project. Those
f a c t E m b e covered under the following headings:
1 Managerial philosophy of planning.

2 Functions of planning:
(a) launch planning
(b) control planning.
3 Level of detail in a plan.
4 Planning the planning process:
(a) project objectives and strategy
(b) evolutionary life-cycle of project planning.
Managerial philosophy of planning

&t%gs~ential
-- that-all-those_-involved in aprqject-should
adop~manageria!philospphpwhEli azcspts that planning
is an integral part of their function. AlthoughThis appearsfo
be simiIar to such conceptsas patriotism, motherhood etc..
lip-service is often made to the importance of planning, and
it is sometimes more difficult to achieve than one would
think
Knowledge and application of the techniques of planning
are not sufficient by themselves to ensure the effective
planning and management of projects. Just as important if
A
25
not more so, irrespective of the techniques used, are: the

adoption of this managerial philosophy: the gaining of the
understanding of those involved of how, and when to apply
the various planning techniques; just what they can and
cannot achieve; and how to build and use a plan to manage a
project.
Formal project planning is very difficult! Planning
requires systematic analysis, the ability t o l o o k . & s n d
~ ~ d e r s t a n d i n ~ o ~ t h ~ a i go
t i into
e ~ ah project
at
and
t h e E n i & e l a t i o n s h i p ~ I t also-incFeaFing6'-involvfs a
kFo=dge =nithe ability to use modern quantitative and
computer-based techniques of planning. Finally it requires
above all imagination. creative abilitv and thinkine about
intangibles. ~ h ; for
s many people to use planning effictively,
it implies a change in philosophy. Planning is a philosophy,
not so much in the literal sense of the word, but as in attitude
or way of life. Planning necessitates a determination to look
ahead constantly and systematically as an integral part of
the management process, and to take decisions and allocate
resources based on a n integrated plan of action. Thus
effective project planning is difficult to carry out and puts
much more emphasis on a manager's conceptual skills, than
does the normal day-to-day management of operations.
& - R , e ~ g a n c eto formal planning often occurs because many
r n a n a g e r s l i a ~ ~ e r i i n c e -ifo,&Tddfindit
f
difficult to
d o the necessary thinking through-required. They-are far
m z e anoxem-azaging roctine operations, thanin thinking
syiGemitical5-&out the future. Some are also reluctant to be*
i,nrivFd in plznning, a s they di~lire2iF7@~Xdis@la~
tlieir
plans or lack of them toxenim manxgeKent. They prefer to
w o X b ~ t l i e ~their
o f pants, in a 'laissez-faire' situation,
and not to be subject to control action, or post mortem
analysis of their planned work
As a result of this, and managerial lack of experience in
formal planning, together with the time, effort and knowledge of techniques required for it, many firms employ
planning specialists, or set up separate planning departments
to carry out this function. Unfortunately this in itself is
insufficient to achieve effective planning. 1-ble-to
just
- 'graft'-on-formaQlanning
to-a n-organisation; there has
p
L
-
26
ADVANCED PROJECT MANAG EMENT
&&eac_ha ~ d ~ i l o s o p and
h knowledge
y
of
everyone concerned with the project. A separate planning
department often leads to a concentration of techniques,
inadequate recognition of the human factors involved and
the adoption of an incorrect attitude towards how plans
should be used.
Unless the people responsible for preparation have done a
thorough and logical analysis, the plan will not be realistic.
Similarly, unless the people responsible for executing the
plan, work in accordance with the information outlined in it.
nothing will have been achieved by planning. Advanced
techniques are probably necessary forthe effective planning
of all but the smaller project, but they are in noway sufficient
to ensure its success. The human factors are as important, or
more important than the techniques involved.
One of the main-reasons-for thefailure of many plans is
t h a m f h a v e ~o~beennmadeebYYthetc~rrectpeop1e.
The
intEduction of the staff planner and a separate planning
department has meant that often either planning is left in the
specialist's hands, or that he can only get limited cooperation from line managers.
When plans are made by a staffplanneractingon his own,
they will have a high probability of being unrealistic, as it is
just not possible for the planner to be fully informed. and
able to exercise critical judgement on what is said to him in
the many areas and functions of a project. More importantly
his plans will be unlikely to be carried out, as human nature
being what it is, most managers have their own way of doing
their job, and dislike being told in detail how to do it.
%~ffective PI-I.I.
9 the n
iv-of
all those
cmrned,b ~ t h ~ e n s u r g h ~ r r e c t n e s s - o f ~ ~ ~
obtain t h e ~ rcommitment-to it. If this participation and
commitment is not achieved, plans will not be realistic and
they will not be used.
A prerequisite for this participation is that the line
manager himselfwants to, and is able to, plan. Therefore one
ofthemajorqroblems~encountered
-in forma-g
is
gettiqgthe-functional line managers-to accept planning as
an integral part of their job, and to do the planning for their
am-of -responsibility;
This in turn requires top management
--
27
support and an understanding of just what planning can

and cannot do, and a correct philosophy of how to use a
plan.
It is thus essential to the success of planning, that on no
account must it be leTt entirely in the hands of a separate
planning department or planner. The planning specialist
should only have staff responsibility for planning and
functional managers should be responsible for the development of their own plans, as planning is a basic function of
any manager. The staff planner should only assist the
manager by expressing his plans and thoughts with the most
appropriate planning techniques, and to integrate them into
the overall project plan. The manager in charge must do the
actual planning for his area of responsibility. He is the only
one who has the knowledge, the power to implement the
plan and in the end, the one who is responsible for and must
live with the outcome.
h h e project tanager,
-- or plann~ngdepartrnent'ss~le~is
the
e x x o n of l e a d e r s h i p j n ~ e s t a b l i s h i n g ~ ~ ~ ~ ~in~ 1 a n ~ a n d
th-rdination-of
every manager's planning effort; the
proms-iaris_of
---- haniIs21ia-Echniques To express the plans,
and the provision of information-to-base plans on, the
intezatGniof th2>l%ning effort, and the educzibii ofmnagemei;t
- - - - -& _ E i t h ~ ~ % i p
n h~ i~l ~ ~ ~-.
p h y-a n d
techniq~~e~thatis,the~art_and~science
of-planning. T i e specialist planner should act as a 'catalyst' in thc planning
process.
The use of sophisticated methods does require specialist
assistance and knowledge. It is difficult for a functional
manager to develop all the skills and techniques for advance
methods of planning and, probably more importantly, to
have the time to apply them. Nevertheless the functional
manager should be aware of the broad principles of the
methods which are used by planning personnel on his
behalf. Although the detailed work of planning might be
taken out of the manager's hands, he is still ultimately
responsible for it, and it must be'his' plan. That is to say, the
project manager with or without specialised planning assistance will provide the framework and the motivation for the
overall project plan, and the managers in charge of each
28
ADVANCED PROIECT MANAGEMENT
sub-organisation must do the planning for their responsib i l i t i e c ~ h project

e
management goupcan provide some of
the necessary inputs, interrelationships, technical assistance
and carry out the necessary paper and quantitative work
Functions of planning
Project planning does not simply mean the logical sequencing and scheduling of the work involved, which is what
many people understand by it. It includes all the work
necessary: to take the'seeds' of a project from its conception
and launch it into the full-scale flowering of an operational
project; to provide the base lines for the ongoingmanage
ment and contrXXThe projea, and to guile it throughout
i t i l ~ s ~ I i 8 n Tand
e ~eEntKdo
i
not quitepork as
pEnntTd. Thus it_&an>e_split-into-two_separatefu~~ns,
or
stxges7.-
A 'launching', or organisational function, in which

planning sets up the project organisation, sequences
and schedules the work, the manpower and the
money in an efficient and effective manner to achieve
the objectives of the project, and establishes the base
lines for control of the project. This is essentially a
decision-making and resource-allocation stage.
2 A 'controlling' function, in that as an integral part of
the control cycle, planning is used to re-allocate
resources, make further or change existing decisions
to steer the project throughout its life as change and
the unexpected occurs.
Launch planning
The amount of information for planning, that is the project
seed, available to the project manager when he takes over
responsibility will vary considerably depending on the
industry and the individual project- from the most elemental
information to an almost fully defined project specification,
for example:
29
1 Elemental This can be an idea, an instruction from

the company's board of directors, or a telex from a
research group etc.
2 Partially defined This can include such projects as a
research and development project, followed by production, which by definition is only partially defined
at the start of the project. It can also include projects
carried out on a cost plus basis, which are characteristically only partially defined.
3 Fully defined This is typically any fixed price tendered
contract, with the most complete examples being in
the building and civil engineering industries with a
complete set of drawings and bills of quantities.
The amount of information available at the 'start' of a
project will also vary for the different organisation's project
managers involved on the same project. For example, the
client company's project manager may take over a project at
the elemental information stage, say a research study,
develop it into a partially defined stage, and go out to tender
to a management contractor. The management contractor's
project manager will then take the partially defined project
information, develop into a fully defined project and go out
to tender with that. However the project is defined and no
matter how much or little information is available to the
project manager when the project becomes his responsibility
he and his planners must take what information they have
and launch the project.
This launch involves far mor~ttbaasimglybujldinga-bar
c h a r ~ f ~ a ~ = w ~ ~ ~ ~ i n ~ l u ~ d ~ o l l ~ w i
pKn=gareZs%hich
-
--
must be integrated and consistent:

-
_+-
Planning the managemen< i.e. the project organisation.

uC Y
:/,'
The project must be broken down into its component
3 parts, the work involved identified and the responsi* -w@ bility for carrying it out allocated, i.e. who does what.
, $ d ~ h i s may be a relatively simple process for the
'
"smaller project, but as they become larger and more
:>.
complex with many organisations and people
30
involved, this area of planning takes on greater

importance.
Planning the work, i.e. the conventional plan.
This is the sequencing and scheduling of the work
and the generally recognised area of planning
familiar to most people.
Planning the manpower, i.e. manpower or resource
planning.
The manpower resources required to complete the
project, that is managers, technologists and manual
labour. must be planned as rarely can the work be
planned in isolation from the manpower resources.
Few project schedules are not limited by the resources
that can be made available to complete them.
Planning the money, i.e. budgeting.
The time-phased expenditure of money must be
planned. that is budgeted; a budget is merely the
expression of the work and manpower plans in terms
of money.
The baselines for any project control system are
the project plan and a time-phased budget based on
it. The project estimate of costs, and perhaps manhours is not suificient on its own to permit effective
control of costs. Once an estimate of cost, which will
probably be roughly based on how the work is going
to be carried out, is completed, it must be expanded
into a time-phased budget of expenditures based on
the project plan. The cost estimate and the project
plan must therefore be integrated into a time-phased
budget, with cost centres based on planned activities,
each having budgeted expenditure for every time
period involved. This involves considerably more
work than simply preparing an estimate, but except
on small projects, if it is not done, financial control
and thus overall management control is lost, and the
project manager cannot manage his project.
Planning the control and information system.
The control and information system for the project
must be planned, even if it is only the allocation of
cost codes to cost centres. However on the larger
31
project, planning the control and information system

is a major area of planning which strongly influences
all the above areas of planning and is critical to
effective management and control.
Control planning
There is both a second function and second stage to project
planning and if the above can be termed launch planning,
then the second stage can be termed control planning. Once
a project is launched the plans and budgets.-proaiicFd
provlde-ffIiXaXe linesagainst which p?ogress-and_performance c a n-b e-compared. Without these base lines, there
w3urdTe no systematic way of assessing performance or
identiFying troublespots, and they are essential to the establishment of the control cycle, as shown in Figure2.2 inwhich
control planning is an integral part
As~s~o~a~roject.executio~commences,dataiscollected,
and in particular the measurement of inputs. and-output~.~
TKemcEp%tZ aKeZs6iitialIy resource inputs of manpower
a-moTey+ and outputs are--progress on the project as
determined byYcompleted~activities, events-achieved and
r a t ~ o f f o u t p u tThe data is then analysed by various
apprTacXei30 produce information as to performance and
troubles, which is fed into the control planning function.
Without this control cycle and second stage to planning,
particularly on the larger or longer projects, launch planning
can become ineffective.
One of the great fallacies held by many managers is
expecting events to work out as planned! The only instances
where this is likely to occur are in the short term, or where
almost every pertinent factor about the future is known, or
where there is inevitably, a straightforward, simple sequence
of events with known resource requirements. Even with the
best of planning, very few projects are completed to their
original plan because there is never sufficient information
available at the start of a project to plan accurately. Normally
the only time that complete information is available is when
the project is more or less complete.
It is not possible to forecast all the activities involved, to
33
estimate accurately the resources and time required for these

activities and to foresee every eventuality. In fact the activities
being estimated are generally not constants. For example,
the time taken to pour a batch of concrete, obtain delivery of
a piece of equipment, get delivery of a batch of components
through a factory, or to get a decision from the main board, is
affected by many uncontrollable variables.
There is always considerable uncertainty about events in
the future, and conditions and relationships are always
changing. If any one activity is not completed to plan, then
this may affect in some way all subsequent activities. With
good management and cc-ordination throughout the project
and everyone following exactly the plan, then deviations
may be limited, but how often does this happen? Strikes may
occur, equipment may break down, and I,abour &gs_tbity
mayYbbel o w e r ~ h ~ x p e c t eThere
d . may also be problems in
management efficiency in handling complex projects,
leading to a lack of co-operation and integration. Remember
also that two-ofthe-c-omon-cau_s~of_delay -and-cost
es7alation on projects are design change and delays in
eqZpmFXdeEvery. TKese-mu3 areci future work on the
frqeTf a n x necessitate replanning.
The concept that once having constructed a plan for a
project, a planner can stand back and say to those involved
'carry on, it's up to you', is a complete misconception of the
function of planning and how a plan should be used.
Planning is an integral part of the management process, and
cannot be separated from it. A plan on the other hand is a
management tool, which a planner can build or assist to
build, and this tool should be used by managers involved to
assist them in making decisions, allocating resources, in
organising, directing and controlling the work on the project;
that is, a plan should be used by the manager to help him
manage. It is a tool that is used, not only to assist him in the
initial organisation of the project, but also to assist him in its
ongoing management throughout the life of the project.
In practice the only definite commitments from a plan are
the initial decisions and resource commitments. Every
further planned action and resource allocation is changeable
in the light of future events. A plan must be looked on as a
34
means of outlining the future work on a project so that

initially the best decisions and resource allocations can be
made to achieve the project's objectives. It enables the
manager to examine the chain of cause and effect likely to
arise from these current decisions, in the light of possible
future events and consequent reactions, and to examine the
effects of change.
Even though changes almost always occur to disrupt the
original plan, a plan is still required to evaluate quickly the
effect of these changes on decisions and actions already
taken, and on future resource commitments and actions.
T k c h a n g e s which will always arise in the-future, form
new i n f o r m a t z n - i m a plan, which3khen combirrd
with logical analisis,mlayTiEtate alternative future actions
to-thzse originally_conceived,Withouta p l a G t i s difficult tb
see quickly-whatdhe-effect o f changed-circumstances-have
onpresent and future actions. With a plan, it is much easier
and faster to re-evaKafethe situation in the light of the
change, to determine what corrective action or redeployment
of resources is required to minimise the disruptive effects of
such change.
Thus a plan must be looked on as a dynamic moving
model of the project. and not be 'cast in stone'. It must be
used adaptively and flexibly as a management tool to assist
the manager to make the best decision and allocation of
resources, both now and in the future, to achieve his
objectives. It must not be considered a failure if the original
plan does not work out, but only if planning throughout the
project has not assisted management to manage effectively.
If planning and plans are to be used as management tools,
the planning and control process must be able to react
quickly to unexpected crises and events. In order for a plan to
be used adaptively and as a dynamic model, it cannot be
looked on as a side line or staff activity. Plans must be used
by project management much more integrally and quickly
than previously, and planning must thus be a 'real time'
activity. For planning to be real time, the speed of the
planning activity must match the pace ofthe project work, or
managers will take decisions on a judgemental basiswithout
waiting for planning analysis, and plans will be unrealistic.
-r
35
The planning activity must also be sensitive to the unexpected

nature of changes that occur. One way this can be achieved
is by use of the modern interactive computer-based planning
systems.
Thus planning cannot be considered as a one-off activity.
It is not sufficient to construct it, pin it on the wall and then
stand back, pleased with oneself. Planning and control are
continuous processes, integral with the management of the
project. Planning must always be considered in terms of a
continuous olanninrr and control cvcle. It is difficult to
separate pla;lning and control, each is dependent on the
other. It is useless to make a vlan without havinginfomation
feedback, managerial analysis and control act'on. Within a
short time any plan is usually out of date and both
continuous and periodic reviews are necessary. Without
these, a plan is merely a piece of paperwhich belongs on the
wall as a decoration to impress the uninformed, or more
realistically, soon after its completion, in the wastepaper
basket
Level of detail
%of
m
a
i
o
r
~
~
misapp led is in ugizg them wit_h-aannnapp_rquiatelevelof
d Z E i i k - ~ e j e cpESG&e
t
proved worse than useless,
d u i e m being constructed either in too great a detail, or
in having the activities too aggregated. Thus a major
problem facing a planneLis_what-!evel_offd~-~~~ld
he
r n ~ E e X Z p i ~ e i i $ l a nin.
s
If plans are constructed in great detail, they will be large
and complex, involve considerable work and take a great
deal of time to complete. This means that they will not be
completed in time to influence the early stages of the project,
which require effective planning and control just as much, if
not more so, than later stages. When unexpected events
occur and the logic of the plan may have to be changed, this
will involve considerable time and effort. Also updating the
plan, monitoring progress and re-assessing resource requirements will be major jobs. This makes it difficult to use the
plan as a management tool in decision making, that is, 'real
36
time' planning. By the time the plan is re-assessed, decisions

will have had to be made, and it is too late to do anything
about it, and planning will not be capable of keepingup with
the pace of work and decision making.
Very detailed plans also necessitate vast amounts of paper
work, both in issuing instructions to work supervisors and
in reporting on work carried out. In project conditions, the
more paper work produced, the less likely is it to be accurate.
Supervisors will spend more time on paper work than in
orale and work
supervising and directin
_to-how
- - much
efficiency will be affect
detail managers-supervisars_anddp~ojectOmanagement
i n E i i o n _ s y s t e m s will deal with,before-theaere volume
o-per
work i t s e i f l b r e a k ~ ~ h e s ~ s t edown,
m
- .- -and the
informatGn loses touch with reality.
In addifion, attFmpting b e r r y detail down to activities
measured in hours, or even days on large projects, is counterproductive. It can be done with success for a limited number
of critical activitieson a project, and perhaps is necessary on
maintenance shutdowns. Even on these jobs, frequent
updating is required, delays and upsets occur, and the plan
almost inevitably loses any pretence of mirroring what is
tually going on.
yJnna_dditio~if--$nior managers and supervisors are
comqetent people, they quitejustifiably object to beingtold
to do things in great detail, for example, being essentially
told'to lift your lefthand before your right hand'. They must
be given discretion in their day-to-day work, and in practical
circumstances, there is no question but that they need it to
deal with constantly changing circumstances.
On project work the actual working situation changes
rapidly at the point ofwork, as regards information, material,
work available, manpower available, breakdowns of equipment, changes and delays. The supervisory level of management must handle all these factors and yet follow the logic of
the planned outline of work with skill, adaptability, experience and judgement. No planning system can match the
competent supervisor in planning and scheduling effectively
the hour-to-hour and day-to-day work to cope with continually changing circumstances.
-
37
Finally, a very detailed plan will be very difficult to

understand, inevitably there will be mistakes, information
will be out of date and unreliable, and there is a great danger
that the plan will very quickly be shown to be wrong,
confidence in the plan will be lost and it will not be used.
If actY '
lanned in too coarse a detail, the plan
WG
f u x n c t i < G ~ r E a ~ s i nco-ordinating
g
and
controlling the work on the projectJkople and organisations
will not knowwhat-they haveto do, and when they have to do
it by. Important interrelationships between activities will be
lost,-and manpower planning will be difficult. Control will
be next to impossible, as major deviations in time and cost
could occur within single activities, which would detrimentally affect the complete project, and yetwould not be picked
up by the control system. A project is a complex set of
activities, which involves many interrelationships, coordination difficulties and control problems, and project
plans have to be built in some detail to permit project
management to manage and control the project.
The-basi&tars-whicMeterminehow_detaied
a plan
should be is what-function is it to carry out and-howIusefu1
t h i s d m m f it is for reporting to top management, then it
should be very coarse scaled indeed; if it is for use by work
supervisors, it should show the work in great detail; if it is to
be used for management decision and control then it need
not be as detailed as the works supervisors level, but
certainly a lot more detailed than that used for top management reporting. Thus the detail of the plan should vary with
its purpose, the size and complexity of the project. For
example, a large civil engineering project may be relatively
straight forward, with a few very large activities, necessitating
only a coarse scale plan. Alternatively, a small project
involving the tying in of a new plant extension to a refinery,
during a refinery shutdown, may require finely detailed
planning to complete the project satisfactorily.Thus, highly
critical jobs with many interrelationships may necessitate
very detailed planning.
The planning at work supervisors level should provide
him with a guide to carry out his work, with some flexibility
within an overall framework, which is in effect the plan. For
--
38
example, the supervisor on the construction of an oil

platform module may be responsible for the completing of
two hundred welds by a certain date. If the sequence of doing
each of these welds is critical and interrelated with steel
erection, cranes, riggers, and others, then each weld and the
work on it must be individually scheduled. However, in such
a complex situation, lots can go wrong, and there is a high
probability of the supervisor having to change the sequence,
for good reasons and sometimes also for bad reasons. It is
very difficult for a planning system to keep up with such
changes and be of much use. It can produce the initial plan,
but this soon gets out of date. If it is necessary to plan the
sequence of each weld, then this should be the supervisor's
job and he should construct his own plan, within the overall
framework. Working together with the steel erectors and
riggers, he decides who does what and when. and it is his
plan for his part of the work. If it is not necessary to have
each of these welds done in a particular sequence, then there
is no need to schedule each weld, one single activity or a
handful of them to monitor progress, is often sufficient to
cover them, instead of two hundred, or more activities.
Activities in a plan should be of a size that where
responsibility changes, resource requirements change and
interrelationships occur. then these can be shown. They
should be of a size that permits close monitoring and control
of progress and cost, without excessive paper work so that
things cannot go badly wrong, without being picked up by
the control system. They should be of a size that permits
some flexibility for work supervisors within activities, yet
enforces rigidity ofwork sequence and schedulingwhere it is
necessary. In large projects the problems of how much detail
to construct a plan is partly overcome by using a hierarchy of
plans of differing levels of detail, as outlined later.
Planning the planning process

Planning for a complex project is rarely a matter of sitting
down and making a single plan before work starts. The
planning process itself needs to be planned and organised.
INTRODUCTlON TO PLANNING
39
For example, consider how to go about planning the design,

building and installation of an oil platform and pipeline in
the North Sea. One could not just sit down and do it from
scratch, even if one had done it before.
Given a target date by the company of three years hence,
one needs to evaluate can it be done in the time and how
should the work be carried out - do all the design, building
and construction ourselves, contract out parts, perhaps
contract the whole area of project management? One of the
first cries would be 'I don't have enough information to plan
it in detail'. It might be possible to plan the initial design work
in detail, but it would not be possible to plan the rest until
information came from design and procurement The full
detailed plan of the project could involve from 4,000 to
75,000 activities, all levels of management are not interested
in such a large plan, and nobody could find their work and
see how it interacts.
At the start of all--b u ~ t h e - s m a l l e r q r . o ~ e c ~ i e n t
information is not available to plan
thepmjed.ubt_ail. Its
- ? E ? ~ m a Y Y nhYay+~nJinahsed.
ot
delivery dates are not
yet available, key dates such as Gntract awar2rXre-fiot
k t i o n a host of other factors are uncertain. Even when
this information is available, there is often a need on large
projects for project plans with two or more levels of detail. A
plan which is used for scheduling and control of individual
activities will be too detailed for management review and
decision making. Plans made will vary in detail, depending
on the information available, and the usefulness of the level
of detail to the managers involved.
In addition, the greater the detail, the longer it takes to
produce a project plan. Form--effectiyenessin
project plann-in-ction-must
be taken-very early in the
projectrTherefore, taking th&ime_toproduce a detailed plan
the early stages, will detract from the
-fwtkewhoIezrojeXin
effectiieness of th_e-c_ontrsLi-nthese stages.
'Thu's t E ~ $ ~ i n g - oa fprojecLmustitself be planneddnd
one of the first steps in the planning-of-a-project i s
det~rmhingjustwhat_P~an~.aE_neede_d2_how
they will be
used
detail thev w i l l h e i n , w h a t t ~ e s - w i l l I b e _
- @ p m i e a n d when the plans will-begrepa~ehPlanning
--
40
can be overdone, with some projects requiring less than

others. There is little u s e j n ~ l a n n i n g
t o a depth-which-will
or~f~~~~hic~~~u~ff~ientinfor
not be
a v 6 i i P i which_themanagement_and
~
supenisoo
i n i V m ~ r s t a n-ord follow. Inappropriate planning emphasis leads to a reaction against planning and the
attitude, 'let's stop this paper exercise and get down to real
work'.
Project planning must take into account the following
concepts:
1 Project objectives and strategies.

2 An evolutionary life cycle of project planning.
3 A hierarchy of project plans.
Project objectives and prqect strategy
All too often project planning is considered to be simply who

does what, when. Preceding this operational stage of planning is a decision-making phase, which is unfortunately
sometimes treated in an offhand manner, but which may
'set' the course of the project adversely, and no amount of
operational planning may be able to change this. Thus, this
pre-planning stage is one of the most important stages of the
project's life and one with far reaching effects.
It is in effect the objectives and strategy setting stage of
project planning, in which the overall objectives, in terms of
technical and capacity performance, time and cost goals, are
specifically outlined, and the major strategic decisions are
taken on how to carry out the project.
This project strategy phase involves the analysis of the
possible alternative ways of carrying out a project, and the
probable results, before deciding on a definitive course of
action. The kind of decisions taken at this stage, are whether
to use a major contractor to design and construct the project,
or to design it oneself and use subcontractors on construction. It also includes policy decisions on the form of contract,
the number of direct contractors, their involvement in plant
design, the way in which variations and design will be
controlled and on penalties and incentives.
41
The importance of project objectives and policies cannot

be overemphasised. Throughout the life of the project, the
project manager must compromise between his conflicting
and multiple objectives. He has to balance technical standards against the cost and effect on the completion date; cost
against time and technical standard; and time against cost
and technical standards. To do this logically, to be able to
reason with and be able to stand up to conflicting interests,
and to justify decisions to senior management requires that
he must know the real commercial significance of these
factors. How important is the cost, the time and the technical
performance, and how can he play off one against the other?
Minimum time and cost, with the highest technical performance, are normally incompatible objectives. A clear
understanding of policy is important for everyone involved.
Is it speed at all costs; what is the real value of time saved; is
final cost critical to overall profitability and thus should the
plan be 'poor boy' with a 'tight' design and no spare
capacity?
Completion date versus target date
The forecast project completion date is a highly emotive

factor, that often causes problems and misunderstandings.
It is impossible at the start of a project to accurately forecast
the actual completion date, as there is simply not enough
information available, and what is being forecast is in fact
not a constant. Time to completion is highly variable,
dependant on a thousand factors and decisions taken in the
past, the present and the future. Generally, completion dates
are set with 'rose coloured spectacles' and are rarely met.
Occasionally, completion dates are set loosely by old hands
and when they are met, or beaten, there are congratulations
all round, all quite unjustified.
Completion dates cannot be set at the start of the project;
target dates can. A target date is an objective and a
motivational factor, which may or may not bear any relation
to a realistic completion date. The term 'a project is two
years behind schedule', need not be a reflection on its
managements' efficiency, but may be more a reflection on
42
man's unbelievable optimism. A target date influences the

project's strategy and how the work is to be carried out. For
example, because of a tight target date a firm may have to go
on a cost plus basis to several contractors, rather than
complete the design and agree a fixed price with a single
contractor after competitive tendering. A target date is thus a
date to which project management will endeavour to plan to
finish the project, and it will influence the decisions taken,
the resources required and the resource allocation. It may or
may not be realistic. The actual completion date can only be
forecast with any degree of reliability after considerable work
has been carried out on a project.
In planning to meet a target date, it is always advisable to
allow a 10-20 per cent contingency time allowance. It is
common practice to allow a budget contingency, and it is
illogical not to also allow a time contingency. Thus if desired
target date is three years hence, the date to be used for
planning purposes should be two years and six to nine
months, depending on the stage of definition of the project.
The evolutionary Zqe cycle of projects

Projects and project planning, in any industry, have a
characteristic evolutionary pattern and a distinct life cycle.
Project planning should be a continuous function which
starts, in some form, at the conception of the project and
extends until the project is in satisfactory operation, or is
fully complete.
The following stages can be identified in a project life
cycle:
1 Project development and preliminary engineering.
2 Bidding and contract negotiation.
3 Engineering design.
4 Purchasing and procurement.
5 Construction and, or manufacture.
6 Commissioning.
Contract negotiation and engineering design can vary in
sequence, depending on which stage of the development of a
43
project that the client, or owner, goes out to bid, and the type
of contract involved.
If the planning is started in the project development
phase, a summary plan would cover all six stages of the
project and a detailed plan would be developed to cover the
preliminary engineering and contract negotiation, followed
by the planning of the engineering design work. As work
actually moved into engineering and design, so would
detailed planning move forward to procurement and construction or manufacture, and as work moved into construction, so would detailed planning of the commissioning work
proceed, following what is termed the 'rolling wave' concept.
Though detailed planning must of necessity follow this
rolling wave, because this is how the information becomes
available, the planning for all stages must be integrated. The
work done in any one stage interacts with the work done in
the other stages, and they cannot be considered in isolation.
Traditionally the planning and control ofworkin each stage
was done more o r less independently, with little or no
interchange between stages; each department or company
planning its work as if in a vacuum. This may have been
satisfactory when the engineering and design work and
purchasing could be completed, materials and equipment
delivered, before the manufacturing or construction stages
were started. Unfortunately, many present-day projects have
to be completed in the minimum possible time, and thus the
work in all stages must overlap and, or be integrated to
achieve the shortest possible project duration. I n today's
competitive business world, and with the high cost of capital,
new plant projects must be completed as quickly as possible
to get an economic return.
Thus planning must extend to cover every aspect of the
project from start to finish, and include planning for those
contingencies that can be foreseen. Planning should cover
from project organisation through development, design,
purchasing, manufacturing, construction, to start up and
initial operations. In fact special emphasis must be given to
those later stages as no matter how the design, manufacturing
and, or construction has been handled, one organisation or
company is handing over to another, and any delay is
44
expensive, as almost all the capital funds will have been

expended and are unproductive, until the unit is successfully
commissioned.
Thus despite the difficulties in information, planning
must be integrated over the project life cycle. In particular
there is a logical way of doing both engineering design,
manufacture and, or construction, and they interact with
each other to determine how the job is done. Any inefficiency
or lack of integration in the overall project only becomes
apparent at the final stages, that is, in manufacture or
construction. and it is too late to change things there.
The balancing of design, purchasing, manEfacture and
construction. often involves Dressures or restraints. brought
about bv the resources avafiable at the manufacturin~or
construction stages. On large projects there are often ins;fficient numbers of kev~rofessionalsor trades to meet t h e ~ e a k
demands. The overafi project plan may have to be modified
to take this into account, or manufacturing and construction
will continue as planned, only until the shortage of key
trades begins to bite. From then on, the manufacturing or
construction phases will be resource limited and very little
can then be done at that stage to avoid a delay in completion.
In this situation the theoretical relationship between the
cost of crashing a job, bears little relation to the practical.
Regular overtime reduces productivity very quickly, and is
very expensive. It is often simply worked to enable higher
wages to be paid. This reduction in productivity is not so
marked where work is machine based. Regular overtime
may thus be required to attract labour, but great care has to
be taken to maintain productivity. Short periods of overtime
for specific visible objectives can often achieve this.
In general, a well-run manufacturing or construction
project is difficult to expedite, and at best only a marginal
improvement can be made, at considerable extra cost.
If the shortage of key trades, or long, or uncertain delivery
on key equipment, is anticipated at the planning stage, the
project can be planned to spread the work of these key trades
over a longer period and so reduce the peak numbers
required. This can be done by staging the manufacture or
construction of the various parts, expediting certain areas of
45
the project through the design and purchasing stages, to give

efficient manufacture or construction. It requires difficult
planning and cross planning between units, and intelligent
anticipation to see that the work is continuous in each area,
and continuity of trades and operations is maintained.
Planning the 'other' stages
Most people are familiar and conscious of the need to plan

and control manufacturing or construction. Often less
emphasis is given to other stages of the project and this is
fatal to the project's chance of success. Planning must cover
all stages of the project from start to finish and particular
emphasis must be given to:
1 Project development stage.
2 Design stage.
3 Completion and commissioning stage.
4 Procurement.
In fact the planning and control of procurement is so
important that it should be given as much attention as the
planning of design, manufacturing or construction. Very few
projects are not delayed, or do not have the sequence ofwork
disrupted, by unexpected late delivery of equipment. This
topic is covered in detail in a later chapter.
Project development stage
The project evolution or development phase may stretch for

years or may only be recognisable as a phase for a very short
period. The project may originate from top company
planning, marketing, engineering, research and development or production. Basically the need for, or opportunity
requiring capital investment is recognised by some function
or person in the organisation. As a result a project proposal
is prepared for and discussed by senior management, and
either rejected, held for consideration, returned for further
development or passed to the next phase for further study
and development into a concrete project.
The evolutionary phase falls more into the corporate
46
ADVANCED PROJECT MANAGEMEM
planning area of business than that of project management.

It is only at, or near the end of it, that the project can be said
to be born, and work on it falls in to the executive functional
areas.
Where competitive forces exist in industry, particularly
with new or significantly improved products, the time from
project conception to actual production can materially alter
the overall profitability of the project. Sometimes the time
taken in research and development, if unduly prolonged,
enables the competitor to get in ahead and pre-empt the
whole project.
Thus it is just as important that the research, development
and preliminary design stages of industrial projects be
planned and controlled, as it is for the other more labour
and capital intensive stages. There may not be the spur of
invested capital obtaining no return, but there is the competitive marketing spur of missing a market opportunity,
reducing the time until competition becomes fierce, or not
being able to meet market demands. Time wasted in the
initial stages of the project, when few people are employed,
can be just as important as in the later stages, where many
people are employed. Many project and construction
managers have been faced with the almost impossible task
of expediting a project in its later stages, at a great cost for a
small time saving. when they were aware of delays in earlier
stages, which could have been easily and cheaply avoided.
Often the argument is put forward that it is impossible to
plan the research and development stages. It is true that it is
difficult to carry out detailed scheduling and that time
estimates are very uncertain. Nevertheless, it is possible to
define the WBS, cost accounts and work packages, allocate
responsibility, and thus effectively plan these stages, to
organise them better as a result to set target dates, to react to
delays faster and to control this development stage of a
project.
Once a project is born it moves in to the preliminary
engineering and study phases which cover both technical
feasibility and financial evaluation of the project The scope
of the project is defined, size determined, methods, processes, equipments, costs, sites, etc. all estimated. Various
47
alternatives are considered until a complete proposal is put

forward for the approval of senior management. Many
projects do not go beyond this stage, as the company does
not really know if a project is feasible or worthwhile, until
this stage is complete. The importance of the preliminary
engineering and study phases cannot be overemphasised.
Decision made in it largely determine what comes after both
during design, manufacturing and or construction. During
this phase, which will cost only a small fraction ofthe overall
project, decisions are made which determine 90 per cent or
more of the cost of the project and largely determine its
operating profitability; the wrong process chosen, whether
to go ahead or not, the wrong size of plant, the wrong type of
equipment, whether to contract the work out or not, the
timing, the speed of the work, etc. Wrong decisions on these
could lock the company in to unprofitable investment for
many years.
Design engineering
Traditionally the main emphasis on planning and control

has been on the manufacturing and construction stages of a
project, but in recent years more and more emphasis has
been given to the planning and control of the design
function. In practice it is critical to the success of a project
that just as much time and effort should be given to the
planning and control of design as is given to the planning
and control of manufacturing and construction.
Effective planning and control of design is difficult but
not impossible. There are a great many complex interrelationships, particularly wit6 information requirements
and designers do not like to be made to commit themselves
to task completion dates, or be put under time pressures.
Nevertheless it is essential that work responsibilities should
be clearly defined, information requirements between functions determined and that the design work be scheduled
with work packages established with visible milestones
identifying achievements along the way. There is often the
need to break work packages into great detail and establish
'inchstones' to-monitor progress. This is where the structuring
48
ADVANCED PROJECTMANAGEMENT
of the design work using the project work breakdown

structure and its extension to the establishment of cost
accounts and work packages, acting as pseudo subcontracts
can be of considerable benefit. Each designer would have
his or her own work package commitments and each work
package would have its milestones and inchstones to monitor
progress. The responsibility matrix would then identify who
is responsible for what. Using this method, design work can
be planned and controlled.
Often the pattern of these design interrelationships is
characteristic to the company or industry, and standard
planning modules or outlines can be constructed as described
in a later chapter. The design process for the characteristic
project is studied and the interrelationships determined.
This is then given a format to build a plan of the engineering
design for all projects of that type. For example, Figure 2.3
shows a summary of an intermediate scaled bar chart forthe
design of a petrochemical plant which is characteristic of
process plant projects. Figure 2.4 shows the S charts described
in a later chapter, which give a measure of the rate ofwork on
these activities.
The major part of the process design has probably been
completed in the development and preliminary engineering
phase prior to these figures, and the bulk of the detailed
engineering design will start when the company gives the
final go-ahead, or a contract is awarded, which is week 0 on
this bar chart. The initial work in the design engineering
phase will be on three main areas:
1 Completing the area layout for the plant
2 Specifying and ordering all vessels and equipmemt
(i.e. the bill of materials - BOM).
3 Completing the engineering, electrical and instrument flowsheets.
A key milestone in process plant design is the completion

and acceptance by all concerned of the preliminary area
layout This is necessary before work can start in earnest on
the other functional areas, that is, structures, foundations,
underground and aboveground piping and before the rate of
work can accelerate on instrumentation and electrical work
ACTIVTW
--------------*-------.------.------.-.-.4--.--+..--4..-.-...*.1
2
BOV-ESSELS
WM-mBE 1x01.
10
BOM-AIB elm.
BOI-mAT
+~~IMW(XX
BoPI-?w5
*%A
LATOWS
*Urn
rXUWT(Xm
+UXULU(~~X----'XXXXIXXXmXXXXLUKK------------1000(--
TNST.IVCrY.1
EU(I.SNCIN.1
I1
UIG
I2
C.S.
PIP.ENCN.L
FIELD '1P.E
13
rnrn.*EII
14
16
ALLOY P.PIP.FY.
C.I.EWOP 'II.6
ALLOY SKI? P.E
I7
B I U R . S I E E L WC..
18
19
20
TOI?ID.INCTN.
W7-INSIIUTL1
15
2,
22
?3
'
2
I
XX(mm---
r(
UIC PIP,E"cr!,.*
32
33
II
M*-mAL.BUILD
Im--Fwl).'UI.L
10
------------------------
xx..
.
.
XXXV.IlXULVXaLliXXXWUKWXUX---XU
U--
sum
FLTP.
XXXXXXXXXX-
MWD.FUCIN.
BOM-ULM P W T .
ALLOY l/F P I P . E .
C I S EH.PLB.'Ts.P..
CIS SH.PAI.PII.B.
IL.Sm.WY.B/EVU
ELeCL.E"clN.2
Umu.
xx-
~ Y O ! X a - -
.
+
r.
A1
W-ALLnY
42
ALLOT SRIF P.W.
45
BOH-DAIAIINC
hb
80%-IWIIILATIOII
IS
"lr. PIP.EYC.2
'6
BW-INSTbUT.2
47
'8
BOW-FO(IIMO.IUT.2
BOX-ALLIIP E/F I.
A9
BOH-CIS SIP PIPE2

B@4-CIS PIPEIU~T.
5"
5,
52
801-C.S.PM
SHIV
--
- ~ * X X X X - - - r(XXX(x-m~ ~ U T ( U Y X ~ X X ( X ~ X ~ O C ~
BW-PIPE
39
nxXXXX
mX-..
IBIT.IYC~.2
IYSI.PIP.ORA,J.
INST.WP.DR*V.
11
18
.
.
n----
--
SEUXXImrLm,xZxXXYX----
~XXWXXUX
xrmXXmrLm,xa(-W--.-
=c---
=---
r--
t
t
---------------.----+P~
Bw-*L",v
PW*T
QUI.PZACT.DSO1-3
--
.
.
aayyxyyy-----mXXY.YUU----
19
3"
IS
16
37
.
.
xyxwumm
I10
8 0 H I U I C PIP&W*T
I
C.S.P/FLW.PT'.E
m-PUt&UOD.
Bo" 9TRLICI A.
WP STRUCI B
WH-EIM.STPPR.
26
27
28
60
*rU3aXLIm---
~Y-DWE
10
SO
Y)
-yuyyIyyyx=
+I(UXIXIXUIX------.XXX*XXX-*XU--
IOY-TOWR
6
8
9
YI
20
rl--
.M
.-I
--------
a---
Figure 2.3 Typical petrochemical project design plan

49
70
INTRODUCTION TO PLANN~NG
51
One of the critical activities which must be started as soon as

possible is piping design. Piping drawings may represent
33-45 per cent of the total drawings required on a process
plant, but work cannot reasonably start on piping design
until 8 to 12 weeks after the start of detailed design, as is
shown in the figures. Piping design cannot start until a
preliminary layout is completed and vendor drawings are
available showing details of equipment and vessels. Similarly
it is inefficient to start work, prior to receipt of vendors' data,
on such activities as the design of foundations or structural
supports.
The area layout establishes firm locations for: vessels equipment - buildings - structures - routings and evaluations of main pipe runs - vessel orientation - platform
locations - evaluation of all equipment - size of structures
and buildings - general routing of underground piping and
electrical conduits - clearances for maintenance personnel.
These initial activities are not separate from each other,
but involve a continuous flow of information backwards
and forwards. For example, in order to prepare a layout,
drawings of vessels are required so that one of the first jobs
started may be to prepare simple line sketches of every
vessel. This togetherwith design specifications, are passed to
purchasing who obtain vessel drawings from the manufacturers involved. These go to the design section for approval
and to the layout group to complete the area layout. A
similar process is required for all the major equipment
involved.
Once the preliminary area layout is completed work can
start in the other functions involved in the project Most of
them have an initial phase which permits the early ordering
of the bulk of the material required with the remainder being
ordered towards the end of each area's functional activity, as
shown in Figure 2.3. All these activities are interrelated and
require a continuous exchange of information. For example,
foundation design and layout is effected by vessel and
equipment details, underground piping layout and structures; piping design is interrelated to vessel and equipment
details of design of structures. Thus though these are shown
as several separate activities on the barchart, there are many
52
micro-interrelationships involved. These can be identified

for projects characteristic to any one company and industry.
Often it is necessary to ensure progress for information to be
made available to interrelated functions before it is accurately defined and different classification of detail are
sometimes used. Thus for exam~le,if foundations are
located approximately, underground' piping runs can be
determined and the connections designed a little later when
foundations are accurately defined and equipment details
available.
The resource requirements for such a plant normally
show a peaked manpower curve, with a ratio of approximately 1.67 to 1.0 between peak and average manpower
requirements. This poses problems when a design organisation is handling more than one project at the same time. In
fact the detail shown in Figures 2.3 and 2.4 were summaries
taken from a single project which had five such units in the
one project. The contractor's schedule was prepared on a
unit basis and only when each function's activities were
consolidated was it obvious that they all peaked at the same
time and something had to give: usually the schedule. Thus
the planning of design and drawing office work has to be
done on a multi-project basis and is discussed in a later
chapter.
Though the actual work on the development and design
engineering phases can be planned and controlled, onevital
aspect is very difficult to control, namely, the obtaining of
key decisions from a company's senior management, or from
a customer. Sometimes it is extremely difficult for the project
manager, acting from a subordinate position, to get them to
make up their mind and take a key decision which is holding
up work Essentially all that he can do is to demonstrate the
consequences of delay and clearly outline the alternatives
available, but often this is not enough. One project manager
stated that in one instance even when told that delay was
costing 100,000 per week, one senior manager could not be
moved to make up his mind quickly. Not only does this have
direct financial consequences for the project, but indirectly
it can effect the morale and commitment of all involved
from start to finish. This is one area when, if the project
53
manager has the courage, some would say the foolhardiness,

a good project reporting system can be a discipline on senior
managers. Thankfully, today, the planning and control of
design has come to be recognised as critical and most firms
give it the necessary emphasis.
Completion and commissioning

The efficient execution of the final completion, that is,
inspection, clearing up final odds and ends, and commissioning of a new project can be vital to its success. If not
carried out effectively, considerable time and money can be
wasted. At this stage almost all the capital has been invested,
lying there useless, probably incurring interest charges, not
making the profit envisaged, not contributing to the company's cash flow, and missing the market opportunity by
allowing competitors to be established. A 10m project
could be incurring interest of 40 thousand per week The
contractor would also still be maintaining men and supervision on site and not be able to obtain final payment and
holdback Therefore delay at this stage is very, very expensive
for everyone concerned and these final stages must be
planned and controlled as effectively as any other stage, with
attention being given to the following points.
Considerable time can be saved if the phased completion
of the different areas or modules can be planned. If certain
sections can be completed early, inspection, checking and
handover can be commenced much earlier than if it is left
until the end of the complete project. Often the specialised
personnel required are in short supply, and probably the
owner's people involved are limited to the operation staff,
who are much fewer in numbers than the construction staff
available.
Careful planning is particularly critical because there is
normally a large organisational interphase. The contractor's
staff or owner's construction staff are handing over to the
owner's or operation staff and unless the work is dovetailed,
that is, jointly planned, cracks will occur and momentum
will be lost. Planning for phased completion can be commenced long before a project is nearing completion. The
54
more lead time obtained, the better the phased completion

can be obtained. Thus it should not be planned when
construction is 90 per cent complete, but much earlier, and
in fact, commenced, or taken into consideration during the
design and procurement stages.
A major problem at the final stages of a project is the
exception, or'but' lists. During the closing stages of a project
the owner and, or contractor will prepare complete detailed
lists of unfinished work in each area. These must be checked
by all concerned for completeness. The best timing for
preparing exception lists must be judged carefully. If done
too early, the work becomes burdensome, because the list is
too bulky and the expediting effect is lost. O n the other hand,
the task of final inspection entails substantial effort and
cannot be accomplished in a few hours, or even a few days.
However, allowances must be made to give time for omissions
or deficiencies to be rectified.
Another problem near the end of a project is maintaining
pace and momentum. Two failures can commonly occur.
Work progresses rapidly until handover occurs, and then
only slowly. The final checking and commissioning is not
started immediately, because the other organisation has not
planned and organised effectively. Alternatively a contractor
finishes 95 per cent of the work, then withdraws the bulk of
his men and the key high-grade personnel involved, and the
remaining 2 to 5 per cent is left for a minimum number of
men, with lower-grade supervision. Pressure must be maintained on this final 2 to 5 per cent of the job. It needs to be
planned, organised, and controlled with the same pace and
attention as the rest of the project, otherwise it can take
almost literally forever.
3 Introduction to Control
Planning launches a project, and effective planning is

critical to the success of a project, but 'launch' planning is
the dominant function of project management for perhaps
only 20 per cent of the project life cycle. As soon as a project
is launched, control becomes the dominant function for the
remaining 80 per cent of the project life cycle. Indeed after
the launch phase, planning and control merge, through the
control cycle, into one integrated managerial function. Thus
the control function of project management is of equal
importance to the planning function. It is certainly operational for a longer time, takes more effort and is highly
dependent on effective control and information systems.
Control is important because it performs two critical
functions, namely: motivating and monitoring performance,
and a managerial function.
The performance monitoring and motivational function of
control
The traditional, essentially accounting, view of control is
that without an effective control system to monitor performance an organisation, and the people in it, will inevitably
become slack and inefficient. Therefore the time and cost to
completion of a project will always be greater if there is no
real objective measurement of performance and control.
The modem behaviouralistic theories of management and
motivation may query thisview. However organisations and
56
the people in them differ widely and it is essentially true that

some people will always seek an easy life and, if they can get
away with i t will settle down to a low untrying performance
level. On the other hand without effective control information, other people will not know they are not performing
satisfactorily and will thus not apply the additional effort
they could and would do, if they were aware they were not
performing adequately.
The same principles apply to an organisation as a whole.
Where an organisation leads a protected life and is not
subject to competitive pressures, it almost inevitably becomes
fat and slothful. One of the factors common to poor
performing organisations and individuals is a lack of an
effective control and information system and a baseline
performance to compare their own with.
Thus there is always a need for the traditional threatening
function of control and this need is greater the poorer the
performances, namely if you do not achieve a satisfactory
level of performance you will be punished Fear is undoubtedly a strong motivator, but it is a limited one. It
destroys teamwork, leads to distrust, limits initiative and is
not the best motivatorwith competent people. However with
some people, particularly the less than competent, it is often
the only effective motivator.
Nevertheless it is only one motivating function of control,
and control can have a much more positive behavioural
function, particularly with competent people. In project
work, planning and control can be combined to be a strong
motivator to achieve higher performance, by showing people
what they have to do to achieve high performance andgiving
them feedback on how they are performing. Thus planning
and control should be used to promote time and cost
consciousness, in that everyone should know and thus strive
to achieve their personal targets within the overall project
setting. For example, the design engineer should know that
he has twelve weeks to complete his part of the design and
thus be motivated by personal achievement rather than
punishment if he does not do it in that time.
INTRODUCTION TO CONTROL
57
The managerial function of control
The performance monitoring and motivational functions of

control are essential to project work and are the traditional
functions of control. However in project work, there is
another function of control which is critical to effective
management, namely, the supplying of information to
enable the project manager to actually manage the project
and not just simply to administer it.
A project is normally a large complex ever-changing
dynamic organism and it is difficult for mere human beings
to envisage all parts of it in their brains and to know what is
going on everywhere at all times. It is difficult to co-ordinate,
is subject to many changes, and control is essential to enable
the project manager to maintain a grasp of it, that is to
effectively manage it.
Furthermore rarely, if ever, does everything go to plan or
budget. Just as in the military field, a plan only lasts until the
enemy is engaged, so in the projects field, as soon as work
starts, deviations from plan occur. The project control and
information system must tell the project manager quickly
how all parts of the project are progressing, highlight the
problem areas and bring to his attention deviations from
plan so that he, or she, can take action to minimise delays
and costs before they get out of hand, i.e. trouble shooting.
Once a project is launched, control is project management,
and without effective control the project manager has little
influence over the project. It will meander to completion but
it will inevitably take longer and cost more than would have
been the case without effective control. Control is another
name for the ongoing management of the project, and
commitment to effective control must become part of the
managerial philosophy of a project manager.
Managerial philosophy of control
Just as for many people to use planning effectively requires a
change of philosophy or attitudes, so for many people, to be
'in control of' their project or their contribution to it, implies
58
a further change in managerial philosophy. The term 'in

control o f highlights the importance of control. How many
project managers or contributors to a project can truthfully
say they are in control of their project or function? All too
often the project is simply administered and not effectively
managed. If the project manager is not in control, it will be
finished some time and at some cost, but the project
manager will have had very little influence over performance. The people that can really influence performance on
a project are few and far between and they combine effective
planning and control, among other attributes.
The essential attitude to control is that the project manager
actively works at being aware of what is happening on all
parts of the project at all times. It includes a constant
comparison of progress and performance against the base
lines set by planning, a search for trouble spots, and a
commitment to do something about deviations from plan
and planned performance. This involves the formal control
and information systems, a real time approach to control
and control by involvement, ie. you are where the action is,
you are involved in it or talk to the ueoule
who are involved
In it and you knowwhat is happening, when it is happening.
Control makes the essential difference between delegation
and abdication, and decentralisation and fragmentation.
Responsibility for control

Execution is the normal function of the managers involved
in a project, both project and functional and is considered to
be their normal duty, and what they are trained for and
experienced in. Control of technical factors is thus relatively
straightforward, and something engineers and technical
personnel are normally fully competent in. Control of work
in progress, money, changes to design specification, procurement of bought-in items and a host of other factors is much
more difficult than the control of technical factors. This
difficulty is magnified in project work not only because of
the dynamic characteristics of this type of work, but also
because of the fact that the project manager is welding
59
together several companies and, or organisations within one

company, into a single organisation, without the normal
permanent relationships between managers and without an
established information system. Technical personnel involved in a project are thus often not experienced, interested
in, or trained in this area of control, and control of the
project is thus one of the prime responsibilities of the project
manager.
Responsibility for the ultimate cost and time to completion
of the project is one area of the project for which the project
manager has little competition from functional line
managers. This may appear to make a great deal of sense, as
he is the only manager in the organisation at the centre ofthe
project information system who is able to put together cost
and progress for all the stages of the project, in order that
managerial analysis can be carried out and control action
taken. This is particularly so for large projects, where it may
be over a year between the major part of engineering design
work being completed, and the main manufacturing or
construction costs coming in. He is the one manager who
can evaluate the future costs of decisions taken early in the
project, and make trade-offs between time, money and
quality. Thus the normal project manager spends a large
part of his time on the control of his project, which in reality
is merely the ongoing management of a project.
However control is not the sole prerogative of the project
manager; every manager and individual involved must
accept that they have responsibility for the control of their
contribution to the project. They must also have a managerial philosophy of control which involves knowingwhat is
required of them, by when, and personally controlling their
own progress. This concept of 'real' time control i.e. being
aware of how you are performing as you are actually
carrying out your work is the fastest and by far the most
effective method of control. That is, if the individual hits
problems, becomes aware that their contribution is going to
cost more, or take longer than planned, they do something
about it and do not wait for formal control feedback This
action may only be to inform the project manager, but if
planning and control are applied in this way, delay in the
60
taking of corrective action can be minimised. Thus every

activity, function and individual should have their own
control cycle and be committed to a control philosophy for
effective project management.
Why control of projects is often ineffective
Although effective control of a project is vital to its efficient

completion, unfortunately in many projects control is
ineffective, with corrective action being too weak and too
late. It is true that in small short-term projects it is often not
unreasonable to just get on with them as efficiently as one
can and collect the costs of the work for the record By the
time control information would be available, using the
normal accounting systems, the projectwould be completed
and no great cost deviation would normally occur on such
small projects. Unfortunately the same attitude is sometimes
taken, occasionally of necessity through inadequate control
systems, with larger projects, and this can sow the seeds of
disaster.
There are three basic reasons for door control of projects:
1 By definition a project is a difficult endeavour to
control.
2 People problems.
3 System problems.
Projects vary in size and complexity, but by definition most

projects are large complex dynamic undertakings, involving
many organisations and people: Therefore inherently a
project is not an easy thing to control. As soon as more than
one organisation is involved in a project interface, coordination and communication problems multiply. Simply
keeping track of technical problems, physical progress and
expenditure on hundreds, thousands, or tens of thousands
of activities is a difficult task and anv fault or delav in the
flow of information causes control to be lost
People problems
The same kind of people problems that occurwith planning,
61
also occur with control, but more so! Just as many operational managers have little or no experience of planning
non-routine projects, so they have little or no experience of
controlling a non-routine, ever-changing project. Control of
such an undertaking does not only require a managerial
philosophy committed to control, but it also requires in the
people involved, a well developed analytical ability and a
knowledge of the problems of control.
To control a project requires the ability to gather into one's
brain a wide span of diverse factors, put them together and
identify small signs indicating problems. The ability to do
this in unprogrammed non-routine work is not something
everyone has naturally, and experience, particularly effective
experience, takes considerable time to obtain.
It is thus surprising that, until recently, although project
planning has quite rightly been given great emphasis, and
publications and training courses abound, very few publications and considerably fewer training courses exist for
project control. Training in control was left to the accountants and, in the few industries that use them, to cost
engineers. Project control is concerned with far more than
just cost, and limiting training in control to these specialists,
is like stating that only specialist planners need know about
project planning. Training in controlling a project must be
given to all those involved in projects.
In addition although an effective formal control and
information system is necessary for good control it is
insufficient by itself. It must be combined with an informal
information system and a sixth sense or intuition. Although
the emphasis in the following pages is on formal control and
information systems, it is folly to rely solely on them. A
project manager must have an informal information system
whereby information is passed by word of mouth over
lunch, coffee, personal visits, telephone calls etc. He must
keep up personal contacts so that people will speak to him,
discuss problems, let him know their feelings about things.
This kind of informal information system can often give
earlier warning on smaller signs than any formal information
system.
In this connection, to talk about a sixth sense or intuition
62
is not a joke. A sixth sense will often give an experienced

project manager a 'gut feel' that, although there may be no
formal report, there are problems and things are not going
correctly.This can be rationalised to be the project manager's
subconscious putting together small, seemingly unconnected
facts or feelings to indicate that a problem, as yet undefined,
exists. Although no real rational process of thought, or
analysis, may have taken place, no project manager should
ignore this 'gut feel', and he must always probe and investigate the reasons for i t
The principal form of control is through the organisation's
formal control and information systems, which often leave
much to be desired and inhibit rather than help the
achievement of effective control by the individual project
manager. Thus in addition to the inherent difficulty in
controlling a large complex undertaking the ability problem
and the lack of experience and training of project managers
in the control function, three major factors associated with
an organisation's control and information can lead to poor
control of projects, namely:
1 The firm's control systems are cost accounting
oriented.
2 The traditional variance analysis approach to
'analysis' is inadequate.
3 The 'quality' of some control and information
systems is low.
The-firm's control systems are cost accounting oriented
Often in project work, the only formal control function or
system is in the hands of the project accountant, and to some
people project control and cost control are thought to be
synonymous with one another. Sometimes the situation is
even worse, and the only control and information system on
a project is a cost accounting system. In reality, cost control
is only one subsystem of the overall project control system,
and cannot be separated from control of other factors
without it becoming ineffective. Cost accounting, in turn, is
only one subsystem of the cost control process and is not
63
really concerned with the management function. It is worthwhile to define clearly what is meant by cost accounting and
cost control in project work.
Cost accounting systems are traditionally concerned with:
'accumulating actual cost, ensuring these costs are properly
allocated against the operations that have to support them,
verifying that the work is actually being carried out, is billed
correctly, the legal requirements of accounting and establishing the capital account for depreciation purposes'. Standard
accounting procedures and systems of checks are employed
to ensure the validity of such costs. This form of cost
accounting is of course essential, but totally inadequate by
itself as a control system for project management.
Cost control on the other hand can be defined as 'the
maintenance of cost within budget by taking corrective
action where necessary, as indicated by the analysis of
predicted cost relative to budget'. This involves cost analysis,
prediction and reporting, together with management action
to correct deviations, in addition to cost accounting.
However, even this traditional definition of cost control is
not adequate for a project and a different definition of cost
control has to be used. Cost control in a more modern idiom
appropriate to project work can be defined as follows: 'the
correct application of the technical sciences in our industry
to produce the least cost product within given guidelines'. In
reality the project manager is concerned with the financial
management, planning and control of his project, with the
objective of 'minimising the overall costs, subject to satisfactory time and performance constraints'.
Even this approach to cost control has proved unsatisfactory in practice, partially due to the inadequacy of the
historical accounting methods to forecast expenditure and
to control cost effectively, but also due to the incompleteness
of any project cost control system. Cost control systems are
not sufficient in themselves to control project work as there
must be a combination of technical, accountancy and
managerial factors, together with the integration of progress
and cost in the control of projects.
It must be remembered that the objectives of the project
manager, with regard to control of his project, are involved
64
with more than costs. On some occasions he may be more

concerned with minimising time, subject to cost and performance constraints, or in fact, as in the early North Sea
work, project managers may on occasions be concerned
with only minimising time, subject to performance constraints. In these early North Sea oil projects, time was the
prime objective, and cost was secondary and this was a
contributing factor to the 130 per cent or more cost escalation
experienced. One of the problems of multiple objectives is
that normally all of them cannot be maximised, or minimised
at the same time. Some compromise has normally to be
made and one objective chosen, such as the time or cost
objective, with the others looked on as constraints. The mix
of objectives, or balance, may vary from project to project,
but in general the project manager is concerned with
somehow minimising both time and cost, or some balance
of these two factors, subject to a satisfactory technical
performance or standard. He then attempts to control cost
and progress, because by doing so he believes he can reduce
the overall cost and time to completion of the project.
One of the essentials for this effectivecontrol of a project is
the integration of all the subsystems for control. It has been
found essential that the subsystems for planning, budgeting,
information and control of both progress and cost must be
integrated with the work to be done, who is to do it,
organisation structure, and that this integration is the basis
of an effective project management system. It establishes
personal responsibility for progress and cost right down the
line to the place where the work is actually being carried out.
All these systems must go together, as the control of progress
and cost are inseparable, and interact with plan and budget,
with the budget being based on the plan of how the work will
actually be carried out. Unless project managers are backed
up by such effective systems, there will not be an effective
project management, and projects will inevitably cost more
and take longer than necessary. Thus the project management requirements for control are often completely different
from those of a traditional proprietorship accounting system,
and generally need a separate management information
system to the normal accounting system of a firm. This
I N T R O D U C T I O N T O CO N T R O L
65
system must integrate progress and cost. be faster and thus

be more up to date, and cover many more areas.
The variance analysis approach
Variance analysis measures a difference between two factors,
by subtracting one from the other to give a positive or
negative variance. It can be used to show differences
between actual progress and planned, and resources used
against estimate or budget; e.g. common variances in the
control of projects are:
1 Scheduled start versus actual start.
2 Scheduled finish versus actual finish.
3 Scheduled time for an activity versus actual time for
an activity.
4 Scheduled date of milestoneversus actual datewhen
the milestone was reached.
5 Budgeted cost versus actual cost.
6 Measured value versus actual cost.
7 Budgeted manhours versus actual manhours.
8 Budgeted unit cost versus actual unit cost.
9 Budget percent complete versus actual percent
complete.
This method of analysing data to give a measure of performance is used in the following two types of analytical
approach:
1 Cost accountant's variance analysis used in general
industry.
2 Quantity surveyor's variance analysis used almost
solely in the building and civil engineering industries.
Cost accountant's approach to variance analysis The traditional
cost accountant's approach to project control is to compare
the budgeted cost to the actual cost to identify any variance.
A typical example is shown in Figure 3.1.
Quantity surveyor's approach to variance analysis The traditional quantity surveyor's approach to project control is to
67
compare the actual cost to the measured value of the work

completed to identify a variance.
In the quantity surveyor's approach the base line for
performance is not an integrated plan and budget but a bill
of quantities or schedule of rates. This may be a list of prices
for every item included in the project, e.g. for 2,300 cubic
metres of concrete or 3,200 cubic metres of excavation. Each
month a quantity surveyor will physically measure all work
done, price it from the bill of quantities and submit a bill to
the client. This measured value will also be compared to the
costs and a variance calculated as shown for project control.
Theoretically this gives tight control but in practice the
bill of quantities is sometimes not integrated with the project
plan and budget. Unfortunately a traditional bill of quantities is a large, very detailed document and is so structured
that it can only be used for planning and budgeting after a
tremendous amount of restructuring and consolidation to
relate detailed cost items to a package of work Therefore
although it has been recognised for many years that the bill
of quantities provides the basis for effective control of
projects, partly due to the traditional non-management
information-oriented structure and partly due to the vast
amount of detailed calculation required to transform the
bill, it has still to be effectively used for control by many
organisations.
Problems with the variance analysis approach
The variance analysis approach does not adequately make

use of the data available and is being superseded by a more
modem approach, which has come to be called performance
analysis. The variance analysis approach is inadequate for
project control because:
1 It is a historic backward-looking method of analysis.
2 It does not use all the data available effectively and
does not lend itself to an integrated approach with
progress and cost.
3 It does not clearly indicate performance.
68
Backward looking versus forwardlooking Both the cost accountant's and quantity surveyor's approach to variance analysis
look backward in time to tell you what happened in the past,
often the distant past. This is totally inadequate for the
control of a project. The type of information the project
manager must have to control his project effectively is
objective answers to the following three questions:
1 What happened in the past?
(a) Are we on schedule, ahead of schedule, behind
schedule, on the project as a whole and the
individual parts of it? If we have a variation from
schedule, where did it occur, why did it occur and
who is responsible for it? What effect will it have
on the other parts of the project, and what can we
do about it?
(6) Is the work being completed to the budget estimate, or less than the budget estimate, or is the
work costing more for the project as a whole and
the individual parts of it? If we have a variation
from budget, where did it occur, why did it occur,
or what caused it, who is responsible for it, and
what can we do about it?
Variance analysis supplies this kind of information, but not the following, just as essential
information.
2 What is going to happen in the future?
(a) Are we going to be on schedule, on budget, or
over schedule and over budget? That is, what is
the trend of progress and cost? It is vital to
become aware of trends at an early stage or there
is little possibility of influencing them. When
work progress or cost is obviously varying, it is
normally too late to do anything about it. The
secrets of real control are to pick up trends as
soon as they start to occur and to be able to do
something about them.
(b) Thus project management needs to be able to
forecast the final cost, and the completion date
for the project and for the individual parts of i t
69
3 Is the rate of work accelerating or decelerating as it

should?
We are also concerned with monitoring against
plan and budget the rate of build-up of cost and
progress. This is analogous to the acceleration of the
project, just as cost versus time is essentially the
velocity of the project. (It is in this that S charts are an
effective tool.)
Thus the method of analysis used should be
forward looking, predicting what is going to happen
to project performance if present trends continue. It
must also be sensitive enough to pick up adverse
performance almost as soon as it occurs so that
something can be done about it.
Variance analysis does not use the data available effectively to

give integrated control There are three sets of data available
which need to be analysed together to give information for
control:
1 The planned or budgeted expenditure of resources.
2 The actual ex~enditureof resources.
3 The value earned, or more commonly the 'earned
value', by this actual expenditure of resources.
The cost accountant's approach to variance analysis uses
only the budgeted and actual expenditures, and suffers from
the major fault that this only tells the manager whether in
the time period concerned, more or less resources were
expended than budgeted. It does not show whether the
resources expended resulted in the planned work being
completed, i.e. whether you obtained value for money or
earned value. That is, without incorporating the third set of
data, the analysis carried out is almost useless. Without this
additional data you cannot evaluate performance and
progress, and whether the project is going to be completed
within budget. Certainly you need to knowwhether you have
expended the resources planned, but you also need to know
whether they completed the work planned.
For example in Figure 3.1, the budget expenditure of
20,000 on ActivityA may represent the estimated cost of500
70
cubic metres of concrete poured and 22.000 might be the

actual cost of400.500 or 600 cubic metres of concrete poured.
Without knowing how much work was done for the money
there is no way of measuring performance, identifying
problems or forecasting final costs. Say in fact 520 cubic
metres of concrete were poured and the estimated cost per
cubic metre was 40, then an earned value could be
computed, i.e. 520 X 40 = 20,800. This represents what the
work actually completed was estimated to cost and in this
case shows a negative variance of 1,200 instead of a positive
variance of 2,000.
In this example two variances should have been used:
A variance on cost = Earned value - Actual cost
= 20,800
22,000
= 1.200
i.e. this concrete cost 1,200 more than it should have.
A variance on schedule = Earned value - Budgeted
expenditure
= 20,800
20,000
= 800
i.e. 800 more work than planned was completed
These variances are different and more meaningful to the

project manager than the cost accountant's variance shown
in Figure 3.1.
In the quantity surveyor's approach to variance control,
the variance on cost is used but normally not the budgeted
expenditure. Therefore it is not integrated with the plan.
Variance analysis does not adequately indicate performance In
scanning a variance analysis report such as a typical
quantity surveyor's report you have to examine more than
just the variance to determine the problem areas. For
example, both the joinenvork and electrical work may both
have a negativevariance of4,252, but the importance of this
variance differs greatly. In one case it is a variance of 4,252
on 65,668 of work and on the other on 15,656 of work
Thus instead ofjust subtracting one factor from another, it
is actually more meaningful to divide one factor by another
71
to give what is termed a performance index, e.g.:

668 - 0.94
Performance index for joiner work = 65
f 69,920
256 - 0.78
Performance index for electrical work = 15
f19,508
Thus performance on joinerwork is 16 per cent better than
on electrical work Put another way joiners are earning 94
pence for every pound spent, while the electricians are only
earning 78 pence for every pound spent. This one figure then
gives a simple measure of performance which can be used
for cost, manpower and schedule performance. It can also be
used to forecast the final cost and time to completion of a
project. A full description of this performance analysis
approach is given in Chapter 8.
Quality of control and information system
One of the major factors inhibiting the effective control of

projects in many organisations is the poor quality of their
control and information systems. Quality is a factor difficult
to define but there are a number of concrete attributes which
contribute to the quality or lack of it of a control and
information system:
1
2
3
4
5
Timeliness of the control system.

Level of detail.
Data versus information.
Reliability of information and data.
Objectiveness of data and information.
Timeliness of the control system One of the problems with the

control and information system of many companies is both
the reporting period and more importantly, the delay in the
reporting of information after the end of a period. One of the
worst cases was in an organisation where the reporting
period was monthly, and it took four weeks after the end of
the month in question to obtain a cost performance report
on construction projects. This made control almost impossible due to the time lags involved.
72
For example, no reliable assessment on performance

could be made in the first month of a construction project, as
start-up factors confuse the analysis. Thus performance
could only be assessed on the second month's performance
which was not obtained until twelve weeks after the start of
the project; far too late for nine to fifteen month-long
construction projects. Even worse was the fact that if
corrective action was taken, itwould take one month to take
effect and would only show improved performance in the
second month, i.e. in weeks sixteen to twenty. Cost performance reports for this period would not be available until
week twenty-four; if the changes were not sufficient to
achieve the required performance, it was rather late to do
anything else as the project was almost finished
Thus it is critical that the delay in reporting performance
is minimised to one or two weeks after the end of the
reporting period in question. If it is any longer, control
action is too late. In addition it is desirable that in short-term
projects, as above, or at critical periods in longer-term
projects, the reporting period be reduced to one or two
weeks, instead of the conventional month. In the example
above, if the control period were one week and the reporting
period one week, low performance would have been picked
up at week five and the success or otherwise of corrective
action would have been evident byweeks eight to ten instead
of six months into the project.
If the formal control and information systems of the
company cannot provide this speed of reporting, then
project management must resort to informal rough and
ready control systems. Thus the project manager may use
the manual logging of manpower against work item and a
fudge factor to give labour costs. He must use material
commitment logging and stores issues to estimate material
costs, and simply estimate his overheads. This rough and
ready control report can then be corrected retrospectively
using the more formal control reports. It is better to have 90
per cent accurate rough and ready performance reports
quickly, than 100 per cent accurate formal reports too late to
be effective. It is rarely that the company's formal performance reports are 100 per cent reliable.
73
Level of detail The level of detail on performance reports also

greatly influences the effectiveness of a control system. It is
often found thatwhatever the level of detail used in planning
or budgeting, cost information for control can only be
collected using the cost codes existing in the company's cost
accounting system. This can be disastrous and inhibit
effective control, particularly as the cost accounting systems
of many companies are not designed for project work
Consider for example one extreme where only the cost and
earned value were reported for the project as awhole. If there
was an adverse variance, how does the project manager and
the people canying out the work know what parts of the
project are losing money, or whose responsibility it was? Just
as bad may be the case where there is no adverse variance,
but this covers a mixture of profits and losses in various
activities, with the possibility of the activities losing money
becoming more influential in later parts of the project If
more detailed reporting were available, advance warning
would be given and without it the project is doomed to
failure. In addition, no lessons would be learned and there
would be no feedback to estimating, with the effect thus
extending to other projects. If combined with the late
reporting, it is a case of control information being'too little,
too late'.
Of equal or even more importance is the level of detail
reauired to give effective management and control. There is
nopoint in ;eveloping a finely zetailed planning, budgeting
and control svstem which serves no managerial uumose and
will not be operated. The principal purp&e ofpreparing a
time-phased budget is to assist project management to
minimise the cost of executing the project by effective
control. To achieve effective control, a project management
information system is required which works and works
quickly. Having approximately accurate information at the
end of the month is much more helpful to the project
manager, than having accurate information two or three
months after the event.
In conditions most projects work under, you cannot
expect to get the project, design engineering, manufacturing
or construction staff to operate a management information
74
system which returns very detailed information, and often

massive amounts of it, accurately and quickly. Nor, often, is
very detailed information of much use. There are always
variations in cost and schedule on individual small activities.
Time and cost estimates of activities are based on average
performance, and there must be many variations about
these averages. It is the performance of the organisation with
these variations averaged out, which is the concern of the
manager.
Some compromise on the size of control centres for
planning budgeting and control analysis must be made.
They must be large enough not to overload the management
and supervision with too much detail, but small enough to
provide effective control, in that deviations from plan and.
budget are identified before they are the size to adversely
affect the overall project It must be possible for costs to be
estimated and budgeted for these control centres, and the
actual costs and progress to be identified, and collected
without difficulty. Ideally it should be possible to not only
integrate the plan, budget and information system, but also
to establish individual responsibility for these control
centres.
Also it must be possible to consolidate the information
gained at the lower levels to perform an analysis for the
overall project, segments of the project and the organisation
involved in the project This can be done by using the
modern methods of project control.
Data versus information One of the problems that frequently

occur in the management of the larger project is that the
control and information system produces masses of data but
very little information. The distinction between data and
information is important. For example, the data in a
procurement report may be 150 pages of computer printout,
but the information is six lines of printout which states that
six orders are going to be late. Information is the analyses
and filtering of data to draw the manager's attention to
exceptions, trouble spots and deviations. No project manager
has the time to analyse the raw data to produce information,
this must be done for him either by the people producing the
reports, or automatically by computer analysis.
75
Reliability of information and data Unfortunately one of the

bugbears of project control is unreliable and inaccurate data
and information. Two typical examples of this are in
procurement and accounting. How many times have project
managers relied on seemingly satisfactory procurement
reports obtained by telephone progressing, as outlined later,
onlv to find out when the order is due for delivery that it is
act;ally going to be late? How many times has a ~r-oject
been
comvleted and for months afterwards costs keep dribblingin and change a profit to a loss?
Reports which do not take into account such costs are
useless for control and this is particularly evident in material
costs. The cost of materials for a project can be reported at
several stages, each of which affect the reliability of the cost
performance reports:
I When
stage.
2 When
3 When
4 When
5 When
the order is placed, i.e. at the commitment

the material is delivered.
the material is issued from stores.
the material is actually used.
the invoice is paid.
If the material cost is reported when the invoice is paid, you

get the situation outlined above, which is useless for project
control. Control must be based on the commitment stage,
which may be inaccurate in that all the material may not be
used or even delivered, but it is timely and permits corrective
action to be taken if necessary. Therefore control based on
commitments must either be coupled with control based on
issue from stores or at time of use, or it is corrected
retrospectively based on these stages. Control based on
commitments may overstate the expenditure, but it is better
ta do this speedily than under-report expenditure, distort the
true position throughout the project and have 'tail end' cost
reporting.
Objective versus subjective data and information It is essential
that the data and information used for control purposes
should be as objective as possible, i.e. exhibiting actual facts
uncoloured by personal feelings or opinions. One of the
major problems with control is the use of subjectiveestimates
76
of progress, based on opinions. This gives rise to the

common 90 per cent syndrome; i.e. if progress reports are
based on subjective estimates, the project appears to progress
satisfactorily until it is 90 per cent complete. It is then
patently obvious to everyone it is not, and the project
remains reported at 90 per cent complete until reality
catches up with opinion.
Conclusions
The requirements of a good control and information system
are as follows:
1 It must bring to management's attention quickly
deviations from ~ l a n n e ~erformance
d
so that action
can be taken tk correci them, co-ordinate other
activities with them, or to take decisions to effectively
manage the project.
2 It must not only bring these deviations to management's attention, but also highlight them to those
responsible for the activities concerned.
3 It must be able to be used, when combined with the
project implementation plan, to effectively motivate
people to higher performance by:
(a) positively - showing people what is required to
achieve high performance and keeping them
informed of their performance;
(b) negatively - making people aware that if they do
not meet the planned performance, then itwill be
brought to senior management's attention.
A good eontrol and information system must therefore meet

the following standards:
1 It must make use of three sets of data:
(a) a base line level of performance, i.e. plan and
budget;
(b) the actual performance, i.e. progress, costs or
resources used;
(c) the earned value obtained by the use of these
resources.
77
It must use these three sets of data to extract

management information for decision and action
by both variance and performance analysis. This
data must be effectively analysed to present to the
manager information and not just raw data.
It must not only tell you what has happened, but
also what is going to happen if present trends
continue, i.e. it must be forward looking and not
backward looking.
It must be sensitive enough to pick up the beginning
of deviations from plan or budget so that action can
be taken early enough to have an effect
It should therefore produce information quickly,
both for as short a possible time period and as soon
after the end of that time period Therefore although
a four-week period is the conventional measuring
period, weekly and two-weekly periods should be
considered, and the maximum delay in producing
information should be two weeks and more preferably one week
Although the data should be filtered, consolidated
and analysed for senior management, where deviation or problems are identified it should be possible
to trace quickly and easily the root causes of them.
The earned value used should always be objective,
that is factual, and it should not be based on
subjective estimates of progress, except on very
small packages of work.
The data used must be reliable.
It should communicate the information produced
to all those involved and not just senior management.
It should promote time and cost consciousness on
all those involved.
It should be an integrated system covering both
progress and cost.
It should include:
(a) real time control;
(b) control by involvement;
(c) informal as well as formal information, and
(4 'gut' feel.
78
The 'givens' of an organisation's project control and information system often do not meet the above criteria and some
project managers may consider them an unobtainable ideal.
An organisation may be set in its ways and not susceptible to
change, just to manage projects effectively. In trying to
obtain an effective control and information system the
project manager may run up against entrenched opposition
from the accountancy side of the organisation, or often the
computer people, whose favourite 'stall' is 'it will take two
years to reprogramme'.
Whatever the reason, if an organisation's project control
and information svstem do not meet the above criteria and
change cannot be ichieved, it is up to the project manager to
im~rovisea rough and readv control and information
syAem that goes iome way to &eeting this criteria to allow
him to manage his project effectively.
Estimating the Project Cost
Money, whether it is called cost, profit or loss, is of extreme

importance in project management as it is the ultimate
criterion of performance. All good project managers become
part accountants and are involved in the estimating, budgeting, forecasting and control of cost for their projects, that is,
their financial management
The difference between an estimate and a budget is that
an estimate is the more or less simple tabulation of estimated
costs for the work involved on the project. It normally only
has a general relationship to how the work on the project is
to be carried out, for example, it can be based on a Level 1 or
initial project plan. The budget, on the other hand, is a timephased, detailed financial plan, integrated with a Level 2 or
Level 3 plan. Whereas an estimate will normally only show a
lump sum allocated to each activity or cost centre, a tirnephased budget will show the expenditure for each time
period for the activity or cost centre. The estimate and the
detailed project plan are thus the basis for the establishment
of a time-phased budget Once constructed, the budget and
the plan are tools for the ongoing management of the project
and the baselines for the project control system.
Importance of cost estimating
Efficient cost estimating of the cost of a project is vital both

to the client and the contractor. Reasonably accurate estimates are essential to the initial decision on whether to
80
ADVANCED PROJECT M ANAGEMENT
proceed with a project or not, that is, the 'go/no-go' decision.

If project costs are widely underestimated at this stage,
companies can be 'locked in' to uneconomic investments,
with low or negative profitability, undertake the wrong
project, or run out of funds. They are also required by the
client company for the knowledgeable comparison of bids,
where contractors are involved, with very low bids being as
much a cause for concern as very high bids. A very low bid
may imply a contractor anxious to gain the contract and
accepting no profit on his bid; it may also imply that some
factor has been missed out on a bid, or that comers are going
to be cut; or it may simply be errors in the contractor's
estimate. If the contractor underestimates the cost to him of
carrying out a project at the bidding stage on fixed price
contracts, the consequences can be catastrophic for him. At
the very least he can be landed with a loss-making contract,
and at the extreme he can go bankrupt, which in turn can
cause the client considerable trouble and expense. Decisions
on tendering policy, that is, whether to bid a low price or the
estimated cost plus a reasonable profit margin, should be
taken on as accurate an estimate as is possible. Thus the
initial cost estimates, both for client and contractor, must be
as accurate as the information available permits, and must
include allowances for any particular areas of uncertainty.
Accurate estimating is also important throughout the life
of a project to evaluate proposed changes, alternative ways
of carrying out the work and as a basis for effective cost
control. Without a good estimate with which to construct a
budget, there is no reliable baseline for evaluation of
performance. Project management cannot decide whether
work is being carried out effectively, or whether the final cost
is going to ;xceed the originally allocated funds. Accurate
estimates of proiect cost are the basis for decision making and
control of workin progress for both client and contracior. If
these estimates do not bear some relation to the reality of
actual costs, motivation is affected, control is lost and the
project will unavoidably cost more than is necessary.
ESTIMATING THE PROJECT COST
81
Escalation of costs
The project that is completed to its first estimate of cost is the
exception rather than the rule. In reality, it is almost
impossible to complete a project to its initial estimate,
because of many factors, not the least of which is the lack of
information at the early stages of the project life cycle on
which to base an accurate estimate. Because of this the
project manager is concerned with the escalation of project
cost estimates from the day he is appointed to a project, until
long after it is finished. This cost escalation has been defined
as:
'The difference between the final cost, or latest estimate
of final cost, and the original definitive estimate.'
It can also be defined, more simply as:

'The change in estimated cost over time.'
Cost escalation on smaller projects of 10-20 per cent is
relatively common, whereas on larger, longer projects,
particularly those with a high development content or
considerable uncertainty at the earlier stages, the sky is the
limit. For example, cost escalations have been reported of 50
per cent onpetrochemicalprojects, 140 per cent in North Sea
oil projects, 210 per cent on nuclear power stations and 545
per cent on Concorde! With high cost escalation many
companies can be trapped by the'sunkcosts' principle. That
is, some way through a project, cost escalation that makes
the whole project uneconomic, becomes obvious, and yet the
logical decision is to proceed to completion. The money
already spent must be considered as sunk, or written off, and
the decision to proceed must be based on the project's rate of
return on the remaining forecast costs to completion. One
such project was initially estimated to have a 40 per cent rate
of return; halfway through, escalation had reduced this to 12
per cent for the overall project, but there was still a 30 per
cent rate of return on the remaining costs to completion and
the logical decision was to proceed to completion.
Some of this cost escalation is avoidable and it is the
project manager's function to prevent it by effective control.
82
However, in almost all projects, some of this cost escalation

is unavoidable and allowances for it should be included in
the initial cost estimate. Whether this cost escalation is
avoidable or unavoidable, the project manager must be
aware of the factors which contribute to project cost
escalation.
These factors are:
1 Inefficiency.
2 Inflation.
3 The characteristic flow of information on which to
base an estimate.
4 Changes to the contract.
5 Form of contract.
Inefficiency of management, supervision, suppliers or labour

will naturally lead to cost escalation, but this is one of the
factors the project control system sets out to monitor and
control. The other factors are not so amenable to straightforward control, but as they are as critical to cost escalation,
if not more critical than inefficiency, they must be subject to
managerial attention and control.
Inflation of project costs

The cost of materials, equipment, services and labour in
proiect work has considerably increased in the 1980s, and it
is no longer practical to simpiy add a nominal allowance for
this inflation to the bottom of a cost estimate. Neither is it
something that can be ignored in the evaluation of projects,
in the belief that the increase in the cost of the project due to
inflation can always be recouped with the inflation in the
price of the product. Government imposed restrictions orintense competition have tended to limit price rises. Thus
inflation can often have a catastrophic effect on the viability
of capital expenditure projects and must always be discreetly
evaluated and allowed for in the project estimate.
The consequences of this inflation can be different for a
client and a contractor, although it can adversely affect the
cash flow of both. However, no contractor will accept a
83
contract over several years without escalation clauses linked

to the appropriate indices. One example of such a price
adjustment formula is shown at the end of this chapter.
Therefore a contractor can be protected against most of the
worst effects of inflation. Unfortunately this is not a complete
protection as index-linked payments may not apply to all
the payments involved in a contract, such as profit margin,
administrative costs, and retentions and the effects of
inflation in individual projects may be different from that
shown in official indices. Material, equipment and labour
costs, and thus inflation, may vary geographically within a
country, and from country to country, and contracts with
subcontractors and su~nliers
mav involve different inflation
protection terms than idat agreedwith a client These factors
may work to the advantage or disadvantage of the contractor,
but he must be aware of them and make contingency
allowances where necessary.
Payments for the effects of inflation generally lag behind
payments for work done, because of the form of the contract
and delays in the publication of indices, and this increases
the gap between funds in and out and increases the amount
of working capital required. Also as inflation goes up,
interest rates normally also go up and this increases the cost
of Gnancing this working capital. Retentions are also a
problem, as by the time they are paid, often long after a
contract is nominally completed, inflation may have considerably reduced their value.
The client, on the other hand, has no protection against
the ravages of inflation and must forecast inflation and take
it into account in the 'go/no-go' decision. Therefore, using
his time-phased budget and forecasts of inflation over the
projected project life, he must prepare a separate inflation
estimate and budget.
For control purposes, those involved in managing the
project cannot be held responsible for cost escalation due to
inflation, and thus cost control data must be deflated or
inflated to allow comparisons with estimates on a meaningful basis. Prices or rates used in compiling the control
estimate should be those ruling at a particular date, generally
referred to as the 'project base date'. The objective in using a
84
common base date for the project is to establish base date

estimates against which the anticipated final cost and base
date prices and rates can be compared; any variance can
then be identified and analysed as to cause and effect and
corrective action taken as necessary. Using indices forward
from this date permits the control estimate to be corrected
for inflation and to be compared with the current costs of
materials and labour. To achieve this objective it will be
necessary for estimates, tenders, orders, contracts, invoices
and changes to be valued at prices and rates ruling at the
base date, with actual and forecast escalation shown
separately. Since inflation is not under the control of the
project manager, the change in costs due to this cause can be
taken out, leaving the changes in costs which are under the
control of the project manager.
Estimating the effects of this inflation and analysing
progress can often involve time consuming and relatively
complex calculations, and sometimes small computer program or programable calculators are used. The effects of this
inflation can often come as a surprise in long projects
because of the characteristic curve of expenditure rate on a
project Figure 4.1 shows a typical curve for the time-phased
expenditure on a five million pound project over five years,
with the bulk of the expenditure being characteristically in
the second half of the project life cycle, and thus it can be
markedly affected by the cumulative effect of inflation. It
shows the cumulative cost curves for this project with
varying rates of inflation: for example, a 15 per cent inflation
rate over the life of a project would lead to a 47 per cent cost
escalation due to inflation alone. The effects of this would
not really be obvious until the end of the third year of work
on the project because of the time phasing of expenditure.
Simple models on a computer or programable calculator
allow project managers to simulate various inflation rates
and their effect on the project budget and its viability.
Delays in completion can also have a significant effect on
this cost escalation due to inflation. Even without any extra
work involved, delays in project completion lead to significant increases in cost and adversely affect project viability. If
3
Years
Y1
Figure 4.1 Effect of inflation on project expenditure

85
86
ADVANCED PRO JECT M A N A G E ME N T
the project described above takes six years instead of five,

this will lead to:
1 An extra year of general administration, overheads
and supervision costs.
2 Additional equipment costs.
3 Additional costs due to the increased inflation, for
example, at 15 per cent inflation the project would
cost an additional 660,000. This leads to a total of 60
per cent cost escalation due to inflation. It also
makes it more difficult to notice the effects of
inflation unless progress and cost are correctly
linked.
4 In addition the extra year will add at least 550,000
additional interest charge.
5 Finally, but often most importantly in many projects,
there will be a loss of profit from the project for an
additional year.
The characteristic flow of estimating information

It is just as impractical to speakof a single estimate ofproject
cost, as it is to speak of a single project plan. All the
information on which to base an accurate estimate of the
cost does not become available until the project has been
fully defined and planned. Even then the cost can only be
'estimated' or, to put it bluntly, 'guessed', as the eventual cost
is subject to considerable change due to many factors. Until
a design has been fully specified, firm quotations obtained
from suppliers and the manufacturing or construction work
involved fully defined in detail, only imprecise estimates of
project costs can be made. Yet decisions have to be made,
based on these cost estimates, before all the information is
available and a budget based on these estimates established
to permit management control. In practice this is handled by
developing several estimates through the life of a project,
each of which is subject to a level of uncertainty which
hopefully declines as the project progresses and more
information becomes available. Each of these estimates
87
must be linked together by a system which identifies and

controls changes to scope, specifications and estimated
costs, so that it should be possible to trace the reasons for
differences between each estimate.
In addition, an allowance normally termed contingency
or management reserve, must be added to each estimate to
cover the following factors:
1
2
3
4
Errors in estimating.
Minor changes to the design as it progresses.
Minor omissions from the estimates.
Minor field extras, or changes, to compensate for
field conditions, or 'bits that don't fit together' when
it comes to actual manufacture or construction.
5 A management reserve to be used at the discretion of
the proj&t manager for changes required to expedite
the work unforeseen minor contingencies, or other
minor areas of uncertainty.
6 'Normal' variations about the averages which are
used in estimating data.
Traditionally a nominal 10 per cent of the estimated cost

of the project 1s added for these factors, but this is an
oversimplification. The amount of contingency should vary
with the degree of uncertainty of the estimate and thus
differing amounts of contingency should be allowed for
each estimating stage. In addition some parts of the project
may have greater amounts of uncertainty than others and
may merit additions: contingency allowances.
By far the best method of estimating the amount of
contingency to allow, is to use experience of similar projects,
modified for the particular project, together with the best
forecast possible of future conditions. This involves accurate
record keeping and analysis to differentiate all the elements
that went into cost escalation on previous projects. This is
often difficult to do; because the records normally available
are insufficiently detailed or reliable to perform such an
analysis.
These contingency funds should be treated as the project
manager's own money held in a safety deposit vault in a
bank, and only reluctantly released on the basis of formally
88
ADVANCED PROJECTMANAGEMENT
documented requirements.Contingency funds for particular

areas of uncertainty should be documented and controlled
in the same manner. If not required, they should be
eliminated from the project estimate and not looked on as
'fat' available for use elsewhere on the project.
It must be emphasised that these contingency funds are to
compensate for minor changes, errors, omissions andvariations in estimates and are not used to cover major changes of
scope or specifications. These more significant factors
require a change in the estimate and these changes, which
occur naturally during the development stage of a project,
should also change the amount of contingency funds, that is,
if a change adds 100,000 to the estimated cost of the project,
an additional 10,000 should be added to the contingency.
Evolution of a project cost estimate
Thevarious estimates of project cost that are made should be

made at those characteristic points in the project life cycle
where information becomes available to make increasingly
accurate estimates, and these points willvary with the type of
project. The project life cycle can normally be split up into
several phases, for which estimates can logically be made. In
a petrochemical project these phases would be at the:
1 Proposal and evolutionary phase.
2 Preliminary design phase.
3 Design and purchasing.
4 Manufacturing or construction.
Figure 4.2. shows the level of uncertainty associated with all
four phases whereas Figure 4.3 shows an S chart for the
design phase which portrays the normal development of a
design in this type of project.
Proposal and evolutionaly phase
Estimates at this phase are 'rule of thumb', 'order of cost'

estimates, subject to wide variations in outcome. They are
based on experience of similar projects, factored to take into
Time
Figure 4.2 Level of uncertainty of project estimate versus time
91
account increases in cost of labour, material and equipment.

Thus the early cost estimate for a project, particularly one
with a large development content or areas of uncertainty, is
often more an expression of hope than a realistic guideline.
Unless scope and specifications can be changed to 'design to
a price', no one can accurately estimate the cost of such a
project at this early stage. It is not until a specific design has
been firmed up, that a cost estimate can be made with any
degree of reliability. Even then, it can still bewidely out if the
design has major changes, flaws or omissions. Large
development projects fall into a class of their own and no
real estimate of final cost can be made until a considerable
amount of work has been camed out. It is difficult to
estimate the level of contingency funds required at this stage.
At the very least, a 20 per cent contingency allowance should
be made, but this may bear little relation to the outcome of a
development project, o r where undue optimism has been
used in estimating.
Preliminary design phrase

The first real estimates of the cost of the project can be made
at the completion of this stage. This estimate is based on a
preliminary or outline design of the complete project,
tenders for major items and bulk material, and preliminary
estimates for manufacturing or construction. The end of this
phase is immediately prior to the start of the S curve in
Figure 4.3. At this point the first realistic decision on 'go/nogo' can be made. There is still a large amount of uncertainty,
but the areas involvedwill have been identified and adequate
special contingency allowances made for them. A 15 per cent
contingency allowance is prudent forthis estimate; however,
significant changes of scope and specification can still occur
and this allowatice does not cover these changes.
This is the earliest a client company can approach a
contractor, and even then the only form of contract which
can normally be negotiated at the end of this preliminary
design phase is a cost plus contract. The exception from this
is when the contractor bids for a complete turn-key design to
meet specified performance requirements. Any estimate of
92
cost at this stage still involves considerable uncertainty as

shown in Figure 4.2 and a considerable amount ofwork has
to be done to reduce this uncertainty before a reliable
estimate for a fixed price bid can be made.
Design and purchasing phase
In this phase, refinement of the project estimate is almost a
continual process with the level of uncertainty declining as
more information becomes available. The earlier in the
design phase a contractor is approached and a contract
awarded, the more likely there is to be a greater number of
changes to scope and specification, which will change the
estimate and be additions to the contract price. The stage at
which contractors are involved varies with the industry. In
some, such as the civil engineering and building industries,
contracts are awarded with tenders based on more or less
complete design specifications; this limits the number of
changes but does not eliminate them completely. In many
other industries contracts are negotiated at some stage
through the design process, involving greater uncertainties
and the probability of more changes to the design and thusgreater escalation of the contract price. This particularly
applies to high technology industries, where development,
design, purchasing and manufacturing are often overlapped
to shorten the project life cycle, at the risk of greater cost
escalation. If a fixed price contract is awarded for the design
engineering, as well as the manufacturing and, or constnktion, the contractor must either quote a price which includes
allowances for the increased costs that may occur, or he must
carry out a large amount of the detailed design before
quoting a price. Either way it can cost the contractor and
client more.
It is possible to identify various stages in most design
processes at which greater refinement of the estimate can be
made. These devend on the characteristics of the vroiect life
cycle for any p&icular industry. For example, in the design
phase of a petrochemical ~roiect.as outlined bv the S curve
shown in Figure 4.3 these could be as follows:
I At approximately 15 percent through the life of the main
L
<
93
project work, as shown in the S curve, design is 20 per cent

complete, vessels and equipment are 80 per cent specified,
area layouts are 70 per cent complete and most important
equipment is ordered Sufficient engineering studies would
have been carried out to establish that no significant
problems incurring delays or extra costs would be encountered. With this estimate a further project evaluation could
be carried out and the 'go' decision reaffirmed, cancelled or
the project sent back for a re-specification of its scope. This
is the first real point that fmed price contracts can be
reasonably negotiated, but the project can still be subject to
variations due to changes arising out of the remaining
design work If a contractor is involved, he is partially
protected against the effect of changes in design, as these
would be considered as extras to any contract, and a 10 per
cent contingency would be sufficient for him. However, the
client is not protected from the effect of changes and must
make provision for them; possibly a 15 per cenFcontingency
at this stage. In tight bidding circumstances the contractor
has often to forg6 any forn-of contingency.
2 At approximately 25 per cent through the project design
is 40 per cent complete and orders are placed for 70 per cent
of materials and equipment. Vessels and equipment are
completely specified, area layouts are complete and other
areas of design are 25 per cent to 50 per cent complete. A
definite estimate can be prepared at this stage, with a normal
10 per cent contingency. This estimate gives a good 'fix' on
the design and procurement aspects of the cost of the project,
with all vital design decisions taken. Thereafter the principal
uncertainties are concerned with the remaining detailed
design of less important work and the construction stage,
such as site conditions, delays in delivery, management
performance and labour productivity. Design changes after
this stage are particularly expensive, as they involve considerable disru~tionof design and construction work and
rush ordering df material aLd equipment. However, design
is sufficiently far advanced to establish the first reasonable
estimate of-construction costs and a construction cost
control budget, at least for the initial stages on a rollingwave
basis.
94
3 At approximately 45 per cent of the way through the

project, design is 75 per cent complete, construction has
started and another estimate can be prepared to finalise the
construction budget. At this stage it is hard to differentiate
between an estimate prepared as a basis for decision making
and the development of a control budget, from a forecast of
the final cost of the project. One of the reasons for refining
the estimate at this stage is to ensure that the construction
budget is realistic.
Contingency versus job schedule
The amount of contingency allowance can be reduced as the

project progresses, commitments made and the funds
expended. The project manager can use either a graphical
method or a formula to reduce this contingency. One such
formula is
Contingency
= (0% x
expenditure)
+ 5% x (Commitments - Expenditure)
+ 9% x (Funds not yet committed)
+ (1% x Total cost)
A simpler graphical method, based on experience, is shown

in Figure 4.4. Whichever method is used, some funds must
be retained to cover corrective work after the project is
completed.
Cost consciousness in design
Traditionally the emphasis of cost control has been on the

construction, or manufacturing phases of a project. with
emphasis on labour productivity and fundamental management efficiency. This is illogical in the extreme, and attention
to cost control must commence much earlier in the project
life cycle.
The actual distribution of cost between phases of the
project will vary from project to project and industry to
industry, but the general shape of the cost distribution curve
96
is often similar to that shown in Figure 4.5. The principal

phases of a project and the approximate percentage of
project cost incurred in each of them are:
1 Project evolution and preliminary design - 2 per cent
(1 per cent - 5 per cent).
2 Design engineering - 13 per cent (8 per cent - 20 per
cent).
3 Purchased materials and equipment - 55 per cent (40
per cent - 70 per cent).
4 Construction or manufacturing - 35 per cent (15 per
cent - 45 per cent).
This rough distribution of cost between phases has the same
general pattern for the construction and manufacturing
industries, except that for long manufacturing runs, the
proportion given to evolution and design engineering is
reduced. In some projects, operation and maintenance costs
must also be taken into consideration by one or more of the
parties involved, that is, Life Cycle Costing(LCC), and these
may dwarf the actual cost of the project, for example,
chemical plants, military aircraft systems.
If we consider that a combination of management efficiency of functional management and labour productivity
can vary the costs of each phase by a plus or minus 10 per
cent, theoretically the difference between good and bad
performance in the construction or manufacturing phase is
only 7 per cent of the total project cost. However, if the same
concept is applied to purchasing, the difference between
good and bad performance is 11 per cent, half as much
again; therefore, efficiency in purchasing is as important, if
not more important, than efficiency in manufacturing or
construction. (Unfortunately, in construction work the minus
10 per cent figure is often exceeded.)
Thus management efficiency and labour productivity
undoubtedly can influence the cost of each phase, but the
main influences on what could be called the basic cost of a
project are the decisions made and work done in the early
stages, that is, the project evolution and preliminary design,
and engineering design phases. No matter how efficient or
inefficient the construction or manufacturing phases are,
Figure 4.5 Distribution of project cost between phases
97
98
the basic cost of the project has largely been determined by

decisions taken in the preceding stages. The key decisionmaking phases of a project are the project evolution and the
design phases. The 10 per cent - 15 per cent of the overall cost
of the project expended in these phases basically determines
the cost of the following stages. Control on the other phases
of procurement and construction are more concerned with
performance than with decision making. Figure 4.6 shows
the relative influence of each phase on the basic cost of a
project.
The importance of decisions taken in the early stages on
the cost of the project cannot be exaggerated. The 2 per cent
or so of the project cost spent on the preliminary design
largely determines the cost of the remaining 98 per cent of
the work, and tends to lock the firm into many of the
essential cost elements of the project, which can then only be
trimmed by further control effort. Similarly the 13 percent or
so spent in detailed design engineering can strongly influence
the cost of the remaining 85 per cent of the project. Thus
great emphasis must be given to cost control, or to give it a
more meaningful definition, cost minimisation, in the early
stages and this involves four factors, namely
1 Cost consciousness of all those concerned.
2 Company standards.
3 Design optimisation - 'design to cost'.
4 The early implementation of a n effective change
control system.
Cost consciousness
The preliminary and over-riding need is for the introduction
and maintenance of cost consciousness throughout the
project, but principally for the motivation of the engineering
designers and draughtsmen. The many daily decisions they
make, both large and small, are all reflected in the final cost.
The company's management in general, and the project
manager in particular, have to ensure the training and
motivation of all the personnel involved in a project is
directed towards cost limitation. This is essentially a general
Figure 4.6 Relative influence of each phase on basic cost of

the project
100
management function to ensure that all personnel involved,

have as one of their objectives, not only satisfactory technical
and schedule performance, but also the minimising of the
cost of the project. In project work there is always a definite
objective, with defined technical requirements, desired
completion dates and maximum economic costs, all of
which are influenced or controlled by outside market
requirements. In these three objectives there is room for
trade off and compromise; there are minimum technical
requirements and desirable ones, in which if there is a saving
in time and money, there may be some reduction. Similarly,
it may be possible to allow more time, if it saves money, or
conversely, it may be worthwhile to spend more money to
gain time. These factors should be known by the project
manager and communicated to all the personnel involved. It
should be used by him to effect compromises to gain the
overall balanced success of the project. Thus, the project
team should know the money value of time and technical
performance.
However, one difficulty in control is that the motivation of
most functional personnel is the technical success of their
part of the project. If they have to take a decision between a
design with some element of risk, or a more expensive safe
one, they will inevitably choose the safer unless some other
influence is involved; they will tend to add a little bit here, a
little bit there, a spare pump, a larger pipe size, a higher
quality design, etc. There must be a balanced attitude
between time, cost and technical performance. This attitude
of mind is one of the best cost control tools and cost control
engineers, systems, etc., can only supplement the efforts of
all the personnel involved with the project and design,
purchasing, construction and contractors, and if cost
consciousness is not part of their philosophy then cost
control will be ineffective.
Contractors designing a fixed cost turn-key plant with
control over design and specification, have most experience
of cost minimisation design. Normally a client, or his agent
has produced a preliminary design, or perhaps a full design
and a set of working specifications, which 'set' the cost of a
project. These, in many cases, tend to produce a 'gold plate'
101
design. An engineer's whole philosophy is against producing

a minimum design and specification, that is, a 'poor boy'
design. However, the difference in cost between a 'normal'
design and a minimum design can be 15-30 per cent of the
overall cost.
Company standards
Influencing the cost of many projects is the high standards,
specifications or policies used by many large companies.
Though the project design team may avoid many factors
which lead to over-design, or a gold plated plant, they may
also use without a second thought books of standard
specifications for detailed work which, more or less, lay
down the law 'thou shalt use. . . .'
More often than not these specifications have been
established by design engineers with a view to having a high
quality project or product. These standards can adversely
affect the cost of the project, with sometimes no real
advantage in quality. Standards, specification and policies
should be questioned from time to time as to whether some
cheaper alternatives would not be adequate for the job in
hand and value analysis applied whenever possible.
Very often firms that manufacture for the nuclear industry,
or the military, find it very difficult to compete for commercial contracts, because a philosophy of high quality has been
accepted as a norm from the managing director to welders
on the shop floor. Contractors very often complain of client
companies who insist on certain policies or standards, when
the contractor points out cheaper satisfactory alternatives.
For example, one contractor commented that his client
insisted on a certain type of large centrifugal compressor
and its housing in a specific way, when he had pointed out
that they had just installed a cheaper, perfectly satisfactory
alternative for another firm which cost 20 per cent less.
This emphasis on cost control as applied to cost consciousness, standards and specifications should also be extended
to the drawing office. Through force of habit many draughtsmen use details that lead to high material, construction and
manufacturing labour costs without often being aware ofthe
102
ADVANCED PROJECT M4NAGEMENT
cost effect of their methods or the alternatives available to

them. It is often ironic that a stroke of a draughtsman's pen,
symbolically speaking, can add as much to the cost of the
project as a main design decision or poor management
performance. If the cost conscious draughtsman is given the
job to review finished drawings to look for possible cost
savings, he can normally always find areas in which substantial savings, or minor savings which repeated many
times can lead to major savings, can be found.
Thus there is the need to review the firm's policies,
standards and specifications, together with its drawing
office practices from a point of view of cost minimisation;
that is, a value analysis approach. This is not to pare
specifications to the bone, such that operating costs and
plant life are affected, though this has been done in the past,
but more to establish what are satisfactory cost conscious
specifications and practices, which are adequate for the job,
but not over adequate.
Design optimisation/design to cost
It is generally possible to carry out a cost review of a

completed design and/or set of drawings and find many
savings. Familiarity with a design, plus the motivation of
looking at it from a cost point ofview, instead of principally a
technical point of view, can generally identify areas where
money can be saved by simple changes. This is carried out
bv some contractors and clients. but is more generally found
u;hen 'design to cost' principles'are used. hire is a growing
tendency both in the construction and manufacturing
industriks to specify a cost for a project, or product, prior to
startingworkon it. Design to cost can be formally defined as:
The practice of pre-determining the project, or product,
cost necessarv to achieve a ~rofitablebusiness venture.
and then designing the projkct or product to achieve the
required schedule and performance objectives at or
below this cost.
Essentially the main tool in designed cost (DTC) practice is
the'iteration' of the design. Once the first attempt at design is
103
completed, it is reviewed for likely areas of cost saving.

Generally it is not possible to save money on all areas, and
attention should be placed on those parts of the design with
high cost implications, that is, Pareto analysis; 20 per cent of
the design may account for 80 per cent of the cost. Design
teams can then review or revise the design of these high cost
areas to try to reduce the cost by some target figure, say 10 per
cent. This exercise can be repeated one or more times,
usually with real but diminishing returns at each design
iteration. Of course, this takes time, money and delays the
project completion and normally one or two iterations are
all that are generally possible, but they can save money by
this optimisation of the design. Spending one or two per cent
more of the project cost on the preliminary and basic design
phases can often save 10 per cent on the total project cost. In
particular if life cycle cost (LCC) is taken into account, that
is, the cost of long production runs in manufacturing
products, or operation and maintenance in plant projects,
the expenditure of more time and money in the initial design
stage can be well worthwhile, that is, trading recurring costs
for non-recurring cost.
Often if savings on DTC are critical, that is, in tendering
and competitive conditions to ensure competitive product
prices, or because of budget limitations, changes in project
scope or product specifications may have to be made during
these iterations. For example, in military contracts the last
10 per cent of performance may or may not be critical, but it
can account for 50 per cent of the cost. If these can be
relaxed, substantial savings can often be made.
Form of contract
The relative importance and effect of cost escalation is

different for the owner or client, and for any contractor
involved. The contractor is somewhat less exposed to the
ravages of cost escalation than the client, though this is
influenced by the form of contract. Nevertheless, it may be
more immediately catastrophic and obvious to the performance of a contractor than that of a client A project which
104
costs more than is justified by its return, may be debilitating,

rather than immediately obvious to the client's performance.
Many a company has been locked in to low profitabilitywith
such a project, through poor control, or the sunk costs
principle. Other projects have been saved by the inflation of
product prices. However, inflation adjusted company
accounting will highlight such projects in the future.
The contractor is not generally involved in the stages of a
project with the greatest cost uncertainty, that is, the early
stages. These are primarily the greatest responsibility of the
client company and estimates made at these stages have the
greatest probability of high escalation. The first point at
which a contractor can be expected to make a fixed price bid
is when the project has been sufficiently defined to permit
reasonably reliable estimating. If a contractor is involved
before this, there is little alternative to a cost plus form of
contract.
There are many forms of contract but they are all
variations of two basic types, that is, fixed price, and cost
plus, with the other forms being modifications to overcome
the various disadvantages associated with these two forms.
Emphasis on cost control between participants varies with
the form of contract, as shown in Figure 4.7. With a fixed
price contract, a contractor must have a high emphasis on
cost control, and the client's project manager is principally
concerned with time, standards and limiting changes arising
from other people in his company. The contractor on the
other hand may have mixed feelings about client instigated
changes as, though they disrupt his flow of work, many
contractors make most of their profit from these changes.
With a cost plus contract the client is very concerned with
cost control, whereas the contractor gives it less emphasis;
the more the projectcosts, the more profit he makes. Though
contractors may not consciously follow this line, subconsciously the people involved cannot give cost control the
same emphasis. For example, are they going to risk conflict
with the client's management in order to limit costs, or
simply say the client is always right? Thus the cost plus form
of contract will normally lead to greater cost escalation,
because of the inherent fact that the cost estimate was
105
Client
Contractor
High
Low
Low
High
Cost p l u s
Lump sum
Type of c o n t r a c t
Figure 4.7 Emphasis on cost control
constructed earlier in the project life cycle and thus was

more uncertain, and because of the motivational effects of
this form of contract Thus the forms of contract decided
upon can influence the degree of cost escalation on a
contract.
It is thus essential that the project manager is aware of the
different forms of contract used in project work and their
advantages and disadvantages.
Lump sum form of contract
In a lump sum contract a contractor agrees to supply for a

fixed price a well-defined and fixed requirement, based on
definitive specifications and drawings. The contractor has
the opportunity of making a profit by carrying out the work
more efficiently than his competitors, but at the risk of
106
making a loss if he underestimates, or has trouble with the

project. The owner is able to obtain a minimum price by
putting out the contract to tender for competitive bids.
It can be used when the project design has advanced to a
stage where it can be fully specified, where there is no or
limited new technology or innovations involved. There is
little likelihood of significant changes to design and cost
inputs are relatively stable. When project execution extends
to a year or more this form of contract is almost always used
with a cost escalation clause similar to that described at the
end of this chapter.
Ifthese prerequisites can be fulfilled the lump sum form of
contract is often preferred by both owners and contractors. It
has the advantages for the owner that he is likely to obtain a
minimum price through competitive tendering, and even if
this is not certain, he at least knows the major part of the
final cost of the contract at an early stage; early enough in
fact to cancel it if need be. It simplifies the administration
and supervision of the project for the owner in that he is not
concerned with monitoring the contract's cost. He is only
concerned with ensuring that the work is progressing in an
orderly manner to meet the desired completion date and
with monitoring the quality of the work, that is, is it meeting
the contractual specifications and drawings? The rigidity of
the contract acts as a restraint on the owner's personnel as
any changes or additions to the contract will involve extra
payment to the contractor. However, if there are many
changes to the contract, it will involve considerable additional charges over and above the fixed price as the contractor
has a tied customer, and many contractors make the bulkof
their profit from changes to fixed price contracts.
From the contractor's point of view the lump sum form of
contract has the advantage that it is sometimes possible to
make a larger profit than with other forms of contract. He
also does not have to disclose his costs and efficiency of
working to the owner, and is subject to a minimum of
interference from the owner's personnel.
The disadvantage of this form of contract is that it
theoretically entails a longer project life cycle. The owner
cannot get a contractor working until the project has been
107
fully defined, which also means that the various stages

cannot be overlapped to reduce the overall project time. The
preparation of lump sum bids also takes the contractor
longer to prepare, and thus is expensive to the contractor,
and in the long run to the customer too. There is also a risk
factor involved for the contractor and some contractors in
certain industries where development projects are common
are reluctant to tender for lump sum contracts, at least for the
design and development stages of a project. The owner also
has the risk that where a major project gets into trouble and
the contractor goes bankrupt, he is left with a partially
completed project which may also be entangled with litigation problems.
The motivation and objectives of the contractor may also
not be what the owner would like. The contractor will have
as his prime objective the minimisation of cost and the
maximisation of profit. This may involve trimming the
equipment sizes to the smallest possible and using the
cheapest quality of materials. The owner has to counteract
this by adequate specifications and supervision to ensure
they are being met. In the simple lump sum contract the
contractor does not have any motivation to complete the
project in the shortest possible time. Normally he does not
want to extend the project time as this will usually increase
his overheads and his costs. Nonetheless ifwork is short and
he wants to spin the project out, or if extending the project
will lead to more profit, then it is in his best interest to do so.
Sometimes the completion time is agreed with a bonus for
early completion and a penalty for late completion, but this
does not really solve the problem. The contractorwill always
try to build i; some contingency in time and money to cater
for this, and such clauses are very difficult to enforce. Where
they exist, every time the owner requests a change the
contractor can quite legitimately extend the contract completion time due to the extra work involved, and few
contracts have no changes.
Cost plus form of contract

In this form of contract the contractor is reimbursed on the
108
basis of all direct costs incurred plus a percentage of these

costs to cover the contractor's overheads and profit. The
contractor is limited to a small margin of profit, but he has
no risk of making a loss.
This form of contract has several advantages which in
certain circumstances outweigh its disadvantages. It enables
a contractor to start work before the project has been fully
specified and where project phases can overlap, this can
theoretically reduce the time to completion. It not only saves
time in this way but it also considerably reduces the time
required for contract tendering and negotiation, both for the
owner and the contractor. The cost plus form of contract is
also appropriate to those projects which involve new technology, a great deal of research and development, or are
innovative. It is only in the later stages of such projects that it
is possible to make a reasonably accurate estimate of how
much they are going to cost, and there is little alternative to
the cost plus formula for this type of project. Thus in projects
where the content is uncertain orwhere there is the likelihood
of many changes to scope and specification, the cost plus
form of contract provides a flexible way of employing a
contractor.
The main disadvantage of the cost plus form of contract is
that the contractor has little motivation to control costs as
the more the project costs the more profit he makes; neither
does he have any motivation to meet target dates. In practice
most contractors are concerned to maintain a good reputation, most people want to do a good job and the contractor
must keep in mind the possibility of being barred from
future contracts by the owner if he abuses this form of
contract. However, in a lump sum contract the contractor
normally acts as a constraint on the owner's personnel in so
far as changes and requests for additional work are concerned.
In the cost plus form of contract this restraint is removed and
the very flexibility of this form of contract is a two-edged
weapon which can lead to increased cost escalation.
In addition to this the owner does not know until near the
end of the project what the project is going to cost, and thus
he may be trapped into an uneconomic project. He also has
more expense in that he must employ more staff to administer
109
and supervise the contractor. The owner's company project

management staff must ensure that the contractor is carrying
out the work efficiently, watch out for over-design, give
considerable attention to cost control and supervise the
contractor to see that he is working efficiently and that his
objectives are a fair cost, a minimum time and satisfactory
technical standards. This can be a major source of annoyance to the contractor's project staff as they have the
owner's staff constantly looking over their shoulders, interfering and questioning their decisions and performance.
Other foms of contract

The other forms of contract used are mostly variations and
enhancements of the cost plus form of contract and endeavour
to modify its disadvantages. The simplest modification to the
straightforward cost plus formula is the 'cost plus fixed fee'.
In this the contractor is reimbursed for all his direct costs
and, instead of a percentage fee on top of them, a fixed fee is
negotiated to cover his central overheads and profit. In order
to use this form of contract the scope and specification of the
project have to be reasonably firmed up so that the fee can
bear some relation to the amount of work involved.
Another variation is the convertible form of contract
where the contractor starts work on a cost plus formula on
an ill-defined project and when it has been defined, a fixed
price for the remaining work is negotiated.This gives a quick
start to the project and lets the owner know at an earlier stage
what the total cost is going to be. It does, however, limit the
possibilities of competitive tendering, as, though it is
theoretically possible to use competitive tendering for the
second stage of the work, it rarely works in practice.
Where the design work is reasonably well completed, but
where exact quantities of construction work may be subject
to wide variations, a bill of quantities form of contract may
often be used in some types of projects. In order to use this,
the order of magnitude of the quantities involved such as
excavation or concrete, have to be known as costs vary quite
considerably with the scale of work.
There are also many forms of cost plus contract where
some form of bonus penalty enhancement is made in order
110
to give the contractor some motivation to meet time and cost

targets, one such form is the target cost contract outlined
below. Under this form of contract, as used by a one owner,
the total net cost of the work is paid by the owner as it is
actually incurred and the contractor's profit is paid
separately as a fee subject to upper and lower limits. The
amount of fee is determined by comparing the actual cost of
the works with the target cost submitted with the tender,
which is subsequently updated at regular intervals. Although
the principle function of the target cost is to provide an
incentive to the contractor it also provides a viable basis for
budgetary control of the project.
The bill of quantities submitted with the tender is priced
by the contractor at rates which will cover all cost which
would normally be allowed for in pricing a measure and
value contract, but which excludes all profit. Profit is dealt
with separately by means of a variable fee. The total amount
of the bill of quantities before the addition of the fee
represents the initial target cost. The final target cost is
established by measurement of the work as executed in
precisely the same way as would be the contract price in a
normal measure and value contract.
If the actual cost of the work is equal to the final target cost
the fee to be paid to the contractor would be 4.5 per cent of
the final target cost. If the actuaI cost is less than the target
cost, the basic fee to be paid to the contractor would be
increased by one-third of the difference between the actual
and final target cost, provided that the total fee payable to the
contractor including the basic fee does not exceed 7 per cent
of the final target cost of the project. If the actual cost exceeds
the final target cost the basic fee payable to the contractor
would be reduced by the amount corresponding to one-third
of the difference, and the actual and the final target cost,
provided that the total fee payable to the contractorwould be
not less than 2 per cent of the final target cost of the works.
This can be expressed by a simple formula
Contractor's fee = 0.045 FTC 1/3 (FTC - AC)

FTC = Final iarget cost
where
AC = Actual cost
111
This form of contract can be further modified to give a time

incentive by incorporating a time bonus penalty as shown
by a second form of the formula
Contractor's fee = 0.034 FTC 113 (FTC - AC)

1 2 (TCT-ACT)
ACT
where
TCT = Target completion time
ACT = Actual completion time
This enhancement of the cost plus form of contract

endeavours to give an in-built performance incentive for the
contractor to achieve an actual cost at the end of the job less
than the target cost as presented at the tender stage. This
necessitates running the job in a normal measured contract
form to establish this and target cost. One of the advantages
of this type of enhancement is that it endeavours to establish
a contractor/client management relationship such that most
major decisions are taken jointly and the risks are agreed
and accepted. The owner is then aware of the basic cost and
the deviations and has most of the information to participate
in budgetary control. One basic failure of any such formula
is that in the event of the contractor falling down on both
cost and time, he can immediately revert to his minimum
fee, which is in this case 2 per cent. This gives an adverse
incentive to make the overall cost as high as possible to
achieve a desired return at 2 per cent. Although not by any
means an attractive rate of return, in periods of economic
depression or lack of work, contractors may still find this
acceptable.
Thus the form of contract used has important implications
for both owner and contractor alike and influences their
motivation, their costs and is the basis of their relationship.
In addition it is important, whatever form of contract is used,
that it is well constructed with no 'grey' areas. If a contract is
poorly worded and is capable of biing misinterpreted, there
will inevitably be misunderstandings and disagreements
leading to owner/contractor conflict If a project is to be
successfully completed it is essential that both the owner
and contractor's staff work together in a reasonably har-
112
monious relationship. Either party can easily make life

difficult for the other. but only at the cost of disrupting the
project with the consequences of increased cost for both and
delays to completion. Any form of contract cannot cover all
points and eventualities, and the owner and contractor's
staff must give and take in an atmosphere of mutual trust
and respect for the good of the project. The owner's staff
recognising that the contractor should make a reasonable
profit, and the contractor's staff accepting that they should
give a reasonable performance for this profit.
If any individual owner gets a reputation of being tight
and unreasonable, the inevitable consequences will be that
the contractors who tender will add a contingency for this in
their bids. Thus, though the owner's staff may congratulate
themselves on not permitting any changes and scoring over
the contractor, the initial bids they receive will be higher
than necessary to allow for this.
If a contractor proves unsatisfactory in performance once
a contract is awarded, there is very little an owner's project
manager can do within the scope of that particular project.
The project manager has as the basis for his authority only
the terms of the contract, the contractor's desire for the
maintenance of goodwill and hope for future work. If this is
insufficient then the owner's project manager is trapped into
a project which will cost more and take longer than
necessary, affect his reputation and which will give him a
very troublesome and unpleasant personal life for the life of
the project.
Thus the selection of a contractor for a project is a twostage process, namely, the preparation of the bid list and the
evaluation of tenders, with the first stage being the most
important. The starting point to obtaining good project
performance from a contractor is to choose reliable, competent contractors to ask to tender.
The preparation of the bid list, namely, the list of contractors who will be asked to bid on a project, is thus one of
the most important stages of a project. his is a real decisionmaking phase as thereafter one step inevitably follows
another Gthout any real option. On many occasibns arguments will be made against awarding firm A a contract, even
113
though firm A is the lowest bidder. These are very difficult

decisions to justify on any grounds, as these very same
grounds should have been used to keep firm A off the bid list.
In order to prepare the bid list it is necessary to prescreen
the contractors to be considered in the following areas:
1 The owner's previous experience with the contractor.
2 The contractor's planning and control system.
3 The contractor's technical ability to handle the type
of work involved.
4 The contractor's ability to handle the size of the
project, both from a managerial and financial point
of view.
5 The contractor's present workload.
6 The contractor's key personnel available for the
project.
7 The contractor's policy with regard to cost plus and
lump sum projects.
114
Example o f a cost escalation clause
BEAMA contract price adjustment formula

The contract price shall be subjecttovariations in accordance
with the BEAMA Contract Price Adjustment Formula
(January 1979). If the cost t o the contractor of performing
his obligations under the contract shall be increased or
reduced by reasons of any rise or fall in the rates of wages
payable to labour or the cost of material or transport above
or below such rates and costs ruling at the date of tender, or
by reasons of the making or amendment after the date of
tender of any law or any order, regulation or byelaw having
the force in the United Kingdom that shall affect the
contractor in the performance of his obligations under the
contract, the amount of such increase or reduction shall be
added to or deducted from the contract price as the case
may be, provided that no account shall be taken of any
amount by which any cost incurred by the contractor has
been increased by the default or negligence of the contractor.
Variations in the cost of materials and labour shall be
calculated in accordance with the following formula
(i) Definitions: for the purpose o f this formula:
'Contract price' means: the total sum certified in each

Interim Certificate less the
total sum certified in all previous certificates in respect
of the Plant Works.
'Contract period' means: the period between the date
of acceptance of the Contractor's tender orthe date of
any earlier written instruction
to the Contractor to proceed
with the work and the date of
the Contractor's application
for the individual Interim
Certificate concerned.
115
(id Labour element

45% of the Contract Price will be regarded as the Labour
Element and will be varied proportionately to the difference
between the Basic Labour Cost lndex and the Mean Labour
Cost lndex expressed as a percentage of the Basic Labour
Cost Index. The Labour Cost lndex will be the BEAMA
Labour Cost Index forthe Mechanical or Electrical Engineering Industry as appropriatetoeach section of the works. The
Basic Labour Cost Indexwill be the Labour Cost Index at the
base date; the Mean Labour Cost Index will be the average
of the Labour Cost Indices published during the second half
of the Contract Period.
(jig
Materials element
45% of the Contract Price will be regarded as the Materials

Element and will be varied proportionately to the difference
between the Basic Materials lndex and the Mean Materials
lndex expressed as a percentage of the Basic Materials
Index. The Materials lndex will be the Department of Trade
and Industry Price lndex Order VII (excluding MLH 342) Table 1 - Mechanical Engineering Industries: Materials
Purchased or the equivalent lndex for Electrical Machinery
as appropriate t o each section of the works (1970 = 100)
published monthly in Tradeandlndust~y.TheBasic Materials
lndex will be the Materials lndex lastpublished before the
base date: the Mean Materials lndex will be the average of
the Materials Indices so published during the middle threefifths of the Contract Period.
5 Planning the Smaller Project

and the Basic Tools
Before dealing with the concepts used in advanced project
management, planning and control, it is necessary to have a
full understanding of the planning process, how to plan the
smaller project and the basic tools or techniques used.
Therefore this chapter, together with the Appendix, describes
the straightforward planning of the smaller project, the
techniques available and the modem computer-based
planning systems. Succeeding chapters will extend these
basic concepts into the much more advanced integrated
approach to planning and control, necessary for the effective
management of the larger project, or portfolio of smaller
projects.
The basic steps of planning

Planning is essentially concerned with the 'who does what,
when and for how much'. Therefore, no matter what
planning technique or techniques are used, the following
factors must be determined:
1 The general scope and objectives of the project.
2 The overall project strategy.
3 The activities involved in completing the project
must be identified and specified.
4 Who is responsible for each of the activities.
5 The sequence of interrelationship of activities, i.e.
the planning logic.
6 The time necessary to complete each activity.
7 The resources, normally manpower, to complete

each activity in the time specified.
8 The estimated cost of each activity.
9 When each activity should and can be camed out,
using a planning technique, or techniques, i.e. a
plan.
10 The total time-phased manpower required to carry
out this plan.
11 The available manpower must be compared to the
required manpower and when each activity can be
carried out, perhaps changed to limit the required
manpower to that available.
12 Combine when the activities should be camed out
with their estimated cost to produce a time-phased
budget for the project.
In this chapter it is assumed that the general scope, objectives

and strategy of the project have been determined, as previously discussed.Thus it is concernedwith identifyingwhat
needs to be done to complete the project, i.e. the activities
involved, determining when they should be carried Out,who
is to carry them out and the time phasing of the expenditure,
as detailed in steps 3 to 12 above.
Spectjjtng the activities
The first step in this planningprocess is to breakup the work
required to complete the project into its logical and functional elements, to list these activities and to write a
description or specification for each. This is a simple
exercise for a very small project and it is often possible to set
down and list the activities, after some discussion with those
involved.
With larger projects this identification and description of
the activities involved can often be a multi-stage process. As
a first step, the project is broken down into its major
components, parts or functional activities, and then each of
these activities is in turn broken down into its more detailed
component activities. This process may be repeated one or
more times, depending on the size and complexity of the
project.
118
This explosion of the project into more detailed activities

may be carried out by the project manager or planner, but
more normally the explosion of the larger component parts
is carried out by the manager responsible for them, in
conjunction with and co-ordinated by the project manager
or planner. Although this process is often an ad hoc
judgemental process for the smaller project, it is much more
formalised and systematic with the larger project, as is
described in a later chapter.
Take, for example, a simple project involving the design
and manufacture of a product. The product has two main
component parts, a mechanical sub-assembly and an electrical sub-assembly. In addition to the detailed design,
procurement and manufacture of these sub-assemblies,
there is also some basic design work and the final assembly
of the product. Thus a first stage explosion ofthe project may
break it into four major activities or segments:
1 Basic design.
2 Design, procurement and fabrication of the mechanical sub-assembly.
3 Design, procurement and fabrication of the electrical
sub-assembly and other electrical work
4 Final assembly.
The further multistage explosion or breakdown of these

higher level components or activities into more detailed
activities to be planned is shown in Figure 5.1.
Determining the sequence and inter-relationship of activities
Once the activities are identified and specified the planning
logic needs to be determined, i.e. the sequence and interrelationship of activities. Three basic questions need to be
asked about every activity:
1 What other activity or activities can be started after
this activity is completed?
2 What other activity or activities have to be completed
immediately prior to this activity starting, i.e. what
are the immediately preceding activities?
3 Can any activity start part way through this activity?
Basic design
Basic design
Design
Design-
Mechanical design I
'Mechanical
desi,
11
Dra~f%-Mechanical drauing off ice

off ice
Mechanical purchasing
Mechanical-Purchasing
Mechanical d e l i v e r
Fabrication
Mechanical fabrication
Assembly
Mechanical sssembly I
Mechanical assembly I1
Specify-specify
motor
motor
Purchase Purchase motor
motor
"
'Deliver
Electrical- Design
Electrical
sub-assembly
motor
E l e c t r i c a l design I
E l e c t r i c a l design I1
E l e c t r i c a l drawing off ice
mchasiW5
E l e c t r i c a l purchasing
\Electrical
deliver
Instrument design
Instrument drawing o f f i c e
Instrument purchasing
Instrument d e l i v e r
Fabrication
\Assembly
Assembly I
Assembly I1
P i n a l assembly
Final assembly I
Final assembly I1
Figure 5.1 Multi-stage explosion of project
120
This third question is a check on whether the activities

identified have been exploded sufficiently for planning
purposes. In this example, this check question initiated a
further explosion of several activities to give the list shown in
Figure 5.1. Thus in the example used, a number of activities
are divided into stages to indicate when other activities can
be started, for example:
1 Design of the mechanical sub-assembly- the boughtin mechanical parts can be ordered when approximately 40 per cent of the mechanical design is
com~leted.
2 ~leciricaldesign - when approximately one-third of
the electrical design work is com~leted.the electrical
motors, bought-i~electricalcom~onents
and boughtin instrument components can be ordered, and
instrument design started.
3 Assembly of the sub-assemblies and the final
assembly - final assembly of the component will
only start after a stock of mechanical and electrical
sub-assemblies has been built up and will extend for
some time after the completion of the mechanical
and electrical sub-assembly work
The science of project planning
The science of project planning is concerned with the

techniques used, of which the two most common are bar or
Gantt Charts, and network analysis methods, which are
used under the various names of critical path method
(CPM), critical path analysis, PERT, and precedence diagrams. Another much less common basic planning technique which is used occasionally, is the line of balance. Its
use is normally confined to projects involving small batch
production, for example, military hardware or a housing
estate. These are basic planning techniques, which are used
as tools to outline and determine the timing of the work
involved on a projecf by means of a visual or mathematical
model, that is, what is to be done, when, and then to
communicate this information to those involved.
PLANNING THE SMALLER PROJECT AND THE BASIC TOOLS
121
Bar charts and network methods in their simplest form

are satisfactory for the smaller project, but their effective use
in larger projects, and in multi-project situations, has
problems which necessitate sophisticated methods to overcome them, and the use of supplementary techniques for
analysis and control. The most common of these techniques
are manpower charts, S curves and milestone charts.
These basic planning systems must be used as project
planning and control information systems and must
1 Be able formally to schedule all the necessary activities
on the project in a manner which will permit the
evaluation of actual progress against plan, and
which will identify the interdependencies between
activities.
2 Permit the project staff to build, change and update
the project plan quickly.
3 Be able to be used to control the project.
4 Be able to be integrated at some level, with the
financial control system.
5 Permit the project staff to have immediate access to
the data to use the system as an information system.
6 Above all they must be able to be used for manpower
planning and control.
Before going on to review the actual techniques critically,
it is worthwhile to outline the basic requirements of this
manpower planning and budgeting.
Manpower planning
In their simplest form the basic planning techniques are
used to determine what is to be done when it is to be done;
this is far from sufficient for project planning and control.
They must also be used to determine who does what and
when, as it is farcical to plan the work on a project, without
planning and controlling the resources required, primarily
manpower. Manpower is a scarce, valuable, expensive and
sometimes troublesome resource, which must be actively
managed.
122
This involves forecasting the manpower requirements

from the initial plan, comparing these requirements with the
manpower available or which can be recruited, and also
comparing the pattern of manpower requirements to good
practice. These factors often impose restraints on the project
plan and require it to be modified to take into account such
factors as
1 Resource/manpower limitations.
2 The avoidance of undue fluctuations in manpower
levels.
3 The overall resource pattern and its effects on
productivity.
There are often limits to the manpower available forwork on
a project, both skilled tradesmen and professional engineers.
In such cases the project is said to be resource limited, and
this becomes a major factor in preparing the project plan
and may dominate all other factors. It may make nonsense
of the critical path and considerably extend the project life.
In such cases recognition of the constraint at an early stage
can lead to better balancing of work and reduction of peak
manpower needs to avoid these constraints.
The naive preparation of a project schedule based on the
basic planning techniques may lead to considerable fluctuations in the number of skilled men required. This may have
been possible in days gone by, when management could
hire and fire at will, but these days are gone. The effect of
such a policy today leads to low productivity, poor industrial
relations and, when skilled tradesmen are in demand, a
reluctance to be employed on such projects. Ifthe manpower
is kept constant, when requirements vary, it is not only
expensive but also generally leads to low productivity and in
turn to poor industrial relations.
The skilled manpower required should be built up and
run down in a planned manner, and violent fluctuations in
the numbers of each trade or profession required should be
avoided. This necessitates scheduling the activities on a
project so that the cumulative requirements are less than, or
equal to the manpower available and that the numbers
required do not fluctuate unduly. Thus manpower planning
and work scheduling interact, and both are necessary for

effective project planning. This will not guarantee higher
productivity and good labour relations, but if not carried
out, there will be low productivity and poor industrial
relations.
The normal way of portraying this manpower plan is to
construct a manning curve, that is, a graph of manpower
against time, and this is derived from the bar chart or
network plan of the project. It is made up for the individual
trades, groups of trades, as shown in Figure 5.2, and for the
total manpower for a project as shown in Figure 5.3.
This manning curve can show clearly
1 The estimated number of men required throughout
the job.
2 The actual number of men used on the job to date.
3 The manning correction needed to get back on
schedule.
4 Probable schedule delays if it is not practical to make
up lost time.
5 The rate of manpower build up and run down.
It has the advantage that it is very sensitive to the effect of
undermanning, showing at an early date the necessity for a
changed plan to meet objectives. By applying restraints to
the curve, for example, manpower limits, it permits estimating
a practical completion date. It has the disadvantages that it
does not show the effects of changes or additionalwork, and
that it is only as accurate as the estimate of manpower
required and the associated productivity estimates.
Attention has also to be paid to the overall resource
pattern and its effect on productivity. The shape of the total
manpower curve is important as a guide to manpower
planning and can help the project staff to avoid several
common mistakes. Figure 5.3 shows typical manpower
curves for a project based on the earliest and latest starts of a
network plan. The rate of build-up of the manpower shown
for the earliest start curve is probably too high for the ability
of construction management to absorb these men efficiently
o n the job. Where a large number of men are brought on to
site quickly, the supervision can become overloaded and
No. of men
r-1
I
I
I
I
I
Electrical
design
T o t a l drawing
office
I
L-
I
I
r -I
I
I
I
I
I
I
I
1
I
1
10
20
30
40
Weeks
Figure 5.2 Manpower plan for critical trades

124
Figure 5.3
Project manpower
125
126
productivity will fall. If the latest start curve is followed there

is too steep a descent of the number of men at the end of the
project. All too often the completion of the work is planned
in this fashion, particularly when construction management
attempts to recover from earlier delays. Almost inevitably
this will not happen in practice and the actual manpower
curve will overshoot due to many factors and the job will
over-run. Thus the work should be scheduled, probably
involving a balance between design and construction, to
give a more gradual build up and run down of manpower.
The actual manpower curve in practice may over slip and
tend to give this steep descent at the end of the project, but
this should not be planned at the beginning.
Sharp peaks on this curve should also be avoided as these
will achieve little, considerably increase the supervisor's
loads and thus probably reduce productivity.There is a limit
to the maximum number of men that can be employed on
any size of project without adversely affecting manpower
productivity. Too much reliance should also not be placed
on costly overtime to make up for the extra manpower
required. Working overtime on a regular basis may be
required to attract manpower, but it is an expensive way of
increasing work output. Actual increases in achieved work
may vary from country to country and from time to time but
most contractors have sufficient data to give characteristic
curves of productivitygains due to working regular overtime,
one example of which is shown in Figure 5.4. This productivity loss does not occurwhen work is machine paced, or is
for short periods for specific tasks.
Thus manpower, or as it is generally termed in CPM/PERT
methods, resource planning, must be integrated with the
work planning, or planning becomes a waste of time.
Budgeting
The baselines for any project control system are the project
plan and a time-phased budget based on i t The project
estimate of costs, and perhaps manhours is not sufficient on
its own to permit effective control of costs on a project. Once
Actual m n h o t t r s
per week
wl.
man
Example:
1.
6 days P 12 hour
=
Produeriviry
Effective manhours =
7 2 actual mnhoura p r week

0.88
7 2 r 0.88 = 6 3 . 4
2.
7 days @ 12 hour
84 actual manhours p r vpek
Prodvctivif y
E f f e c t i v e manhours
0.77
84 x 0 . 7 1
64.1
N.B. Curves are only illusrreriv~
Figure 5.4 Productivity with overtime
127
128
an estimate of project costs, which will probably be roughly

based on how the work is going to be carried out, is
completed, it must be expanded into a time-phased budget
of expenditures based on the project plan. The cost estimate
and the project plan must therefore be integrated into a timephased budget, with costs centres based on planned activities,
each having budgeted expenditure for every time period
involved. This involves considerably more work than simply
preparing an estimate, but except on small projects, if it is not
done, financial control and thus overall management control
is lost, and the project manager cannot manage his project.
Figure 5.5 shows a project plan for a very simple example,
and Figure 5.6 shows the time-phased budget for this project
This budget can be constructed manually from a bar chart
and estimated in the same way a manpower chart is constructed. However, the amount of work involved in budget
preparation can be considerably reduced by simple computer
systems, which can take the project plan and estimate, and
automatically produce a project budget for an even rate of
working on each activity. This can then be modified interactively for activities where the rate of work, or expenditure
is not even.
Unfortunately it is not possible to maintain either of the
two critical performance measurement baselines, that is, the
plan and time-phase budget as static baselines. As mentioned
previously, nothing ever seems to go to plan and budget in
project work, and at times there are debates on whether the
original plan and budget should be the baseline or whether
these should be revised to take into account changes that
have occurred. One argument put forward against changing
them is that if the plan and budget are frequently changing,
they are merely being adjusted for inefficiency and they do
not provide valid baselines. The counter argument is that
many changes do occur in project work, and if the plan and
budget are not changed to take some of these into account,
they soon lose all credibility as control baselines.
It must be remembered that the plan as well as being a
control baseline, is a management tool to assist the project
manager to make decisions and to allocate resources to
achieve his objectives. It must be used in an adaptive way to
Arrow diagram
B-7 weeks r\ E-LO reeks

I9.Oa)
U E12.000
A-5 weeks
c7.000
1-6 weeks r \ J - 1 weeks

r8.om
Vrs,om
C r i . Path
c r i . Path
cri.
Path
Cri. Path
cri. Pllh
Bar Chart
11
1
2
Aetiviky
A
B
F
G
8
9
H
I
10
0
0
0
0
0
0
0
0
0
20
10
Men
30
XllXX
YXXXXXX
XXXXXXXX----KYXXXXUXXKXXXX
XX
.
.
XXXXXKXM Y XXXXU
----
XXXX
Figure 5.5 Project plan for simple example

129
40
50
131
remain a real time aid to project management. It is also true

that budgets should not just be changed because costs to
date are higher than budgeted, and progress is less than
planned. They should not be changed merely to compensate
for variances already experienced. The budget is a performance baseline, built up from estimates of how much the
work should cost and how long it should take, if carried out
in a particular way.
If a budget is hopelessly inaccurate, it is useless as a
control tool and it is counterproductive not to change it, as
otherwise no one will pay any attention to it. The budget
after all is a basis for comparing actual costs and when they
occurred, with estimated costs and when they were supposed
to occur. It is thus one way of measuring efficiency and
progress, but primarily it is a means of attracting management's attention to any deviations from planned progress
and estimated cost Thus it is wrong to be dogmatic about
changing a budget and it should be changed whenever it
serves some managerial or control purpose to do so. There is
no real reason not to do so, as analysis can easily be carried
out with more than one budget baseline when using
computer-based systems, by simply using different data
files.
However, in addition to changes in performance affecting
the budget there are a number of factors which quite
legitimately require the budget to be changed During the
life of a project, any budget is subject to an almost continuous
series of minor alterations to cope with changes to schedule,
changes to the scope ofwork and adjustments to compensate
for inflation. These do not alter the essential structure or
base-estimates of the budget, and are not really 'changing'
the budget Whatever changes occur to the budget, it must be
maintained as an up-to-date, valid baseline, against which
to measure performance, and it must be tight, but not
unrealistic.
Thus there are a number of factors which must be
formally incorporated into a dynamic approach to project
budgeting and control, if the budget is to remain such avalid
baseline. The principal factors are:
132
1 The characteristic flow of information in the development phase of a project

2 Changes to the project.
3 Variations in the plan.
4 Inflation.
During the development and engineering design phases of a
project, the project estimate will be refined as more information of increasing reliability becomes available. Thus when
the project estimate is changed, the project budget will
require to be changed to take this into account. No matter
how indefinite the estimate is in the early stages of a project,
a budget should be constructed early in the project life cycle
to show the proposed time-phased expenditure, otherwise
there is no control baseline. It is probable that the estimate
for the early work on the project will be more defined than
that of the later work, similar to the development of different
levels of a plan, and the budget will be a reasonably accurate
control tool for these early stages.
However, even when a definitive project budget is established, it is still not normally possible to keep it static. There
will probably still be design changes to the project, and as
these are made, appropriate changes must also be made to
the budget for this work. Occasionally this involves adding
to the budgets for existing cost centres, at other times it may
be necessary to add new cost centres for which performance
data must be collected.
When work on a project actually starts, changes to the
time phasing of work are almost inevitable, and whenever
the planned schedule of work changes, so must the timephased budget. If an activity is replanned to start later than
originally scheduled, and, or to take longer, then the period
to period time-phased budget should also be changed. This
is a relatively simple change, with the baseline estimate
remaining the same, and merely the time phasing of the
expenditure changed to correspond to the new plan. The
only changes to the total expenditure would be for associated
overheads, allocated on a time-phased basis, and any
adjustment for inflation due to the extended timescale to the
project.
In many projects, inflation makes nonsense of the original

figures used in a project budget, therefore it must be
incorporated in the budgeting process. Performance
measurements can be made on an inflated or deflated basis,
with more or less the same results. Actual costs can be
compared to inflated budget estimates, or deflated actual
costs can be compared to the original budget estimate. It
should make little difference whether the analysis is carried
out on an inflated, or deflated basis, provided like is always
compared to like. Often both methods have to be used for
differing purposes, and for, or by, the different parties
involved in a project. The estimated costs at the base date
prices can be compared with the actual costs deflated to the
base date prices, or the base date estimate can be inflated to
current prices and compared with actual costs. The actual
budget for control purposes would be varying dynamically,
but would still vrovide a valid baseline for control vumoses.
The project bidget is thus normally subject to a Amber of
minorrevisions due to these factors throughout the life of the
project. In addition to this on many p&jects, there often
comes a time, despite these minor revisions, when a major
revision of plan and, or estimate, and thus the budget
becomes necessary. The original plan and estimate may
have been wildly optimistic, or even pessimistic, unexpected
cost, schedule or technical problems may have occurred, or
there may be changes to the project strategy in how the
project is to be carried out Whateverthe reason, the existing
budget is now totally unrealistic, or has been subject to so
many changes that it is extremely complex and difficult to
follow, does not provide a valid performance baseline, and
thus serves no purpose.
In such cases it is necessary to re-plan. re-estimate and rebudget the project, so that the best decisions and resource
allocations can be made in the light of the changed circumstances, and yet control of the project is maintained. This
kind of revision is not undertaken lightly, and certainly
should not be carried out more frequently than once per
year. Even then, the revision of the budget should not
happen more than once or twice in the life of an important
project, and there should be substantial remaining work on
134
the project after the last budget revision.

An exception to this is where there is a change of project
management. When project managers are changed during
long life cycle projects it is prudent for the new project
manager to do a full analysis of his projectwhen taking over.
This is necessary to familiarise himself with his project, but
also to clearly establish 'his' performance baseline so that he
cannot be held responsible for delays and over-expenditures
of the previous management.
Cash flow forecasting
As well as being a performance baseline, budgeting and

control systems perform another function, namely, cash
flow forecasting. With the high cost of capital today and the
problems many companies have, both client and contractor,
in maintaining a positive cash flow, cash flow forecasting
and its management are important activities for many firms.
In the contractor's case, care must be taken to maintain a
balance between funds obtained from the client, and payments made to direct labour, equipment and material
suppliers and to subcontractors. The situation which must
be avoided at all costs is the 'hysteresis' form of cash flow in
and out. as shown in Figure 5.7. The vertical difference
between cash in and cash out must be maintained as small
as ~ossible,vreferably zero or positive, if the contractor is
ex6emely l;cky. ~ h i is
s especially important in foreign
contracts as contractors must also manage their foreign
currency dealings. In many such cases, the profits from this
can exceed the profits from the project; in others, ifthis is not
managed effectively, the losses from foreign currency
commitments can bring about large losses on a project and
can lead to bankruptcy.
The budget is the tool that is used in the initial stages of a
project to forecast cash flow, but as soon as work starts the
budget forecasts of cash flow must be modified by the
analysis of performance. These forecasts are not revised
budgets, and the budget should not be revised simply to
provide a better forecast of cash flow. Thus as work pro-
136
gresses, cash flow forecasting should be based on an

analysis of the existing budget modified in the light of
performance and progress to date.
Bar charts
The oldest formal planning technique in use today is the bar
chart, sometimes termed a Gantt or multiple activity chart.
This technique was developed during World War 1, by one
of the early management consultants, Henry L. Gantt, a
contemporary of Frederick Taylor the father of scientific
management. Figure 5.8 shows a bar chart for the project
outlined in Figure 5.9.
The technique has the advantages that it is:
1 Clear: the chart forms a pictorial model of the
project.
2 Simple: with very little training almost anyone can
learn to construct and use a bar chart, and it is easily
understood by all involved in the project.
3 It can be used to show progress.
4 It can be used for manpower planning.
However, it does have disadvantages which limit its

applicability, namely
1 The bar chart, by itself, cannot show interrelationships between activities on large, complex projects,
and this can lead to problems in co-ordinating the
work.
2 There is a physical limit to the size of a bar chart,
which limits the size of project that can be planned
using this technique, unless a hierarchical planning
system is used.
3 It cannot easily cope with frequent changes or
updating. Each change necessitates the redrawing of
the chart. In smaller projects this can be handled by
having each job represented by a moveable strip, but
this soon becomes impractical as the size of the
project increases.
139
When the number of activities is not too great. the bar

chart is perfectly satisfactory and thus it is by far the most
common technique used to plan and control work on
smaller projects. It is a good communication medium, as the
chart forms a clear picture of the project. It is clearly
understood by all those involved in a project from managing
director to foreman, and with very little training, almost
anyone can learn to build and use a bar chart on the smaller
project It can be used to show progress by drawing a second
bar for each activity to represent when the work was actually
carried out. or alternatively, the percentage completion of
the activity. In Figure 5.8 certain tasks are shown ahead of
schedule, on schedule and behind schedule.
It can be used for manpower planning, or resource
allocation, by marking on it the numbers of different men of
different trades required, in each activity for each time
period. as shown. Total manpower for each trade, and for the
project as a whole, can then be summed up for each time
period and a manpower chart and cumulative manpower
chart, that is, an S chart constructed. On smaller jobs the
activities can be rescheduled with little difficulty, as progress
varies from plan and changes occur. It is also quick to
construct and change.
Its one disadvantage on smallerjobs is that it cannot show
interrelationships, critical paths and the spare time on an
activity, but this can be overcome by using a bar chart in
conjunction with an arrow diagram. This enables float, or
spare time on activities to be determined and manpower
planning carried out by rescheduling activities within their
float to achieve a phased build up and run down of labour
and avoid severe fluctuations. If resources are still limited,
then the finishing date may have to be extended to take this
into account as described later.
Unfortunately as the project size increases, the bar chart
technique has difficulties in handling the number of activities
involved, the mass of data and complex interrelationships
between activities. This can be partly overcome by using a
hierarchy of bar charts, sometimes termed tiered bar charts.
This does enable the bar chart method physically to cope
with a large number of activities in a systematic manner. It
140
does not enable them to display, or take into account

interrelationships and it does tend to given an inflexible
plan, as changing and updating requires significant time
and labour. In addition integrating manpower planning
over the total project becomes very difficult. Thus the bar
chart method, when used by itself has limitations in the
detailed planning of large projects and this leads to the
development of network methods for project planning.
CPM/PERT - network analysis

CPMIPERT or to give these techniques a more generic
name, network methods, are used in project planning and
control under many names and guises. Common names are
Critical Path Method (CPM), critical path analysis, Project
Evaluation & Review Technique (PERT) and precedence
diagrams. The critical path method was developed by E.I. du
Pont $Amours and was first successfully applied in 1958 on
a construction project in Kentucky. At the same time the US
Navy's Polaris special projects office, in conjunction with
consultants, developed a basically similar PERT system.
PERT as originally conceived, used three time estimates for
each activity, one optimistic, a most likely and a pessimistic,
and used a probability distribution to combine these. This is
rarely used today in industrial projects in this form, though
the name is now often applied to the basic network methods,
and it is interesting to note that there is a growing interest in
risk analysis and simulation in project planning using
probabilities in a different manner. Figure 5.9 shows a
network or arrow diagram for a simple project, and Figure
5.10 shows the normal time analysis output from the
calculations involved.
Precedence networks, with an activity on the node or
event. as opposed to the activity on the arrow used in the
other methods, followed in 1962. It is similar and supposedly
slightly simpler in its handling of dummies and ladders, but
the activity on arrow, that is, CPM/PERT still tends to be the
most comon network method used in project planning.
The network method has the following advantages:
1 The method effectively handles the interrelationships between activities on complex projects.
2 It identifies those activities which are critical to
finishing the overall project on time and shows the
spare time, or float, on the other activities.
3 It can handle very large and complex projects with
the present maximum size being between 25,000 and
75,000 activities, depending on the systems used.
4 It can be used with a computer and this permits the
integration of the project management information
systems.
Thus, because projects were becoming larger and more
complex, and their planning more critical to project management efficiency, there was a trend to the widespread adoption
of CPM/PERT in the 1960s and early 1970s. Its use became
accepted as typifying good practice in project planning, and
those who did not use it were considered to be old fashioned,
or even immoral!
Bar charts versus CPMIPERT
Despite the fact that it was fashionable to use CPM/PERT,

barcharts were still widely used, in these years, in both small
and large projects. CPMIPERT never really got established
in the planning of the smaller project, because it was not
really required, its increased sophistication discouraged its
use and bar charts were, and still are, an adequate and
effective way of planning and controlling the smaller project.
Even with larger projects, bar charts were often used as the
primary planning technique, principally because they were
simple, easy to use, project staff were familiarwith them and
they were often adequate for the state of the art of planning.
However, even in the late 1970s, contrary to all the
publicity and literature produced on CPM/PERT, many
companies were actually discarding CPM/PERT and
returning to the use of bar charts, even on larger and
complex projects. For example:
TIME ANALYSIS
---.---------
2
J
4
5
6
7
ACTIVITY
B A S I C DESIOY
MECHANICAL D E S I G N I
ELECTRICAL DESIBN I
PURCHASE UECH.PARTS
MECHANICAL DESIGN 2
S P E C I F Y ELECT.MO1ORS
PURCHASE ELECT.PARTS
ELE C TRI C A L DESI GN 2
INSTRUMENT DESIGN
PURCHASE INSTRURENTS
D E L I V E R UECH.PARTS
MECH.DRAUING OFF.
PURCHASE EL.MOTORS
D E L I V E R ELECT.PAHTS
ELEC.DRAUIIG OFF.
1NST.DRAYIWG OFF.
D E L I V E R INSTRUMENTS
F A B R I C A T E MECH.PARTS
D E L I V E R MOTORS
F A B R I C A T E ELCT.PARTS
ASSEMBLE MECH.l
ASSEMBLE ELEC.1
IIUM~Y
DUMMY
ASSEMBLE MECH.2
ASSEflRLE ELEC.2
F I N A L ASSEIIHLYI
F I N A L ASSEUBLY2
Figure 5.10 Time analysis
5
10
10
15
15
20
20
J
10
15
20
25
30
35
40
One engineering manager from a large US aerospace1

weapons contractor commented in 1979 that CPM/
PERT had fallen out of favour in his company, very
few of the company projects use this technique and he
had not been associated with PERT for five years.
A senior project manager from a large UK chemical
company went out to tender on a 20 million project in
1979, and on a preliminary survey of possible contractors
found that most of them were not using CPM/PERT.
Though in the past the US Department of Defense
enforced the use of PERT on their contractors, they
now state they 'do not require contractors to use any
specific scheduling technique. PERT, Line of Balance,
Gantt and Milestone Charts are all good techniques
which are effective when properly employed'.
Thus in the late 1970s there was a trend in some companies
away from theuse of CPM/PERT, even on the larger project,
back to using bar or Gantt charts for project planning. Even
when CPMIPERT is used, it is often used only as a sales
factor, or because the home office or the client insisted upon
it. This trend away from CPM/PERT has arisen because
many companies had experienced difficulties in using the
technique, and the disadvantages and limitations of CPM/
PERT came to be more widely appreciated. In addition to
the general problems with planning described in the last
chapter, the use of CPM/PERT has the following problems:
1 CPM/PERT is more complicated than bar charting,
and this in turn leads to a lack of regard for human
factors involved.
2 There are many problems associated with the
application of the CPM/PERT techniques on large
projects, and these are not widely appreciated.
3 There were problems associated with the computing
systems used.
Human factors
CPM/PERT is more complicated than bar charting, and
requires a greater degree of knowledge and skill with the
144
technique itself, and with the associated computer-based

support for it. Many managers and engineers lacked this
knowledge and skill, and adequate training was not given to
them. As a result many firms employed specialist planners,
or set up separate planning departments, and this lead to the
problems described in the last chapter.
One of the major advantages of bar charting is that it is a
relatively simple technique, requiring Little training, and the
individual manager or engineer can construct his own with
little difficulty and everyone involved can understand them.
Problems with the CPlWPERT technique
Though a simple CPM/PERT application can be carried

out with relatively little training, realistic size applications
cause many problems in practice, which few, if any, of the
textbooks or training courses on the technique attempt to
cover. In addition to general problems with the art and
science of project planning and difficulties with the computing system, many troubles were experienced due to the
following causes:
1 Difficulties in communication.
2 Difficulties in monitoring and controlling progress.
3 The fact that by itself a network and its associated
time-analysis is not an operational plan or schedule.
4 There was a widespread misinterpretation of the
concept of float
Communication. One of the major difficulties of network
analysis at the supervisory level, and often for others, is that
an arrow diagram is avery poor means of communication.A
large network is difficult to understand even for the specialist,
and much more difficult for the general manager or supervisor, who is actually carrying out his part of the project.
With some training, the manager can easily understand a
simple arrow diagram, or one made up of macro-activities,
such as is used for communication purposes with senior
management. A large and detailed networklooks more like a
map of the heavens, than anything else, and is generally
145
unintelligible to those who have no training. Even with

training, it can still be very difficult to follow and to use.
The time analysis output of network analysis, as shown in
Figure 5.10 is even more confusing, and much worse for
communication down the line. Often this is what is handed
out to front line management and supervision, and when
used for transferring planning into action generally fails.
Even if the person concerned can identify the activity to be
done, all he has is the earliest andlatest start and finish times
for the activity. As far as he is concerned, as long as he can get
the activity done in these times, everything is on schedule.
When he gets the time analysis output for the whole project
the interrelationships of activities are extremely difficult to
decipher.
Control. Simple network based methods have no way of
showing progress against plan to the manager. Normally the
network is time analysed, and possibly resource analysed at
periodic intervals to update it and replan, if necessary, the
activities on the project This can be done at very short
intervals of a day or a week, in such critical applications as
maintenance shutdown scheduling, but more normally on
large jobs the network is updated and re-run once a month.
This essentially produces a new plan or schedule for the
remainder of the work Unfortunately, there is the problem
of relating one network to another. Sometimes no effort is
made to check one run of the network with another and the
project is continually slipping without people being aware of
i t Float is being consumed all the time and managers are
completely unaware of the problems building up.
A network is not an operational plan and schedule. All that

simple network analysis tells you is:
1 It shows the jobs on the critical path with their start
and finish times.
2 It shows the spare time or float on the otherjobs with
the permissible earliest and latest start and finish
times.
3 It shows the interrelationships between activities.
146
It does not actually tell you who does what, when. It is thus
not an operational plan and schedule. A critical path
schedule is really only a loose planning guide. Except for the
critical activities, network analysis produced only a permissive
framework, with some limiting dates that can be used to
guide the scheduling of the project. The bulk of the results
are a long way from being a schedule, that a project
manager, functional manager, or construction superintendent can use effectively.
It has two major omissions:
1 It assumes infinite resources are available and no
recognition is taken of any limitation in manpower,
i.e., no manpower planning.
2 Non-critical jobs are not scheduled.
An operational plan and schedule includes manpower

planning, sometimes equipment planning, and someone
deciding when the so-called non-critical jobs are to be done.
Thus the arrow diagram and time analysis output of a
network are not the final plan and schedule for a project, but
simply the basis, or first step, in preparing an operational
plan.
Problems with the concept offloat. There are major problems
in the use of network analysis with regard to the misinterpretation and misuse of float. For example, the construction
supervision may be planning to delay the start of the first
activities in an area, by not assigning resources until some
time after its scheduled early start, but before its latest start. If
this is done, it will reduce the float available on later
activities on that path, and make the float shown in the
critical path output for later activities unrealistic with
respect to the project schedule.
The concept of float as spare time on a job is avery useful
one in planning by network analysis. Unfortunately, three
types of float are in general use, namely, total, free and
&dependent float and very few people completely understand the meaning of any of these. Total float can be taken as
the spare time o n a n activity, the consumption ofwhich will
affect the amount of float on both previous and subsequent
jobs. Free float can be taken as the spare time on an activity

which, provided that the previous activities have been
carried out to plan, will not affect any subsequent job.
Independent float is the spare time on any activity, the
consumption of which will not affect either previous or
subsequent activities. The tendency today is only to display
total and free float, as independent float tends to be rarely
encountered, and activities having independent float also
have free float. Total float can be taken as the free time basically on a path through the network; that is not on any particular activity but on a sequence of activities- in otherwords,
the macro-activity or consolidated activity. Unfortunately,
very few people, particularly at the supervisory level, understand the meaning of the various types of float. If the word
float is shown against the job, it is assumed to be the spare
time on that job. Almost inevitably this is then used up in the
first half of a project, with the resultant effect that many
activities, if not all, become critical in the second half of the
project. In practice a project can appear to be progressing to
schedule. in that updates of the network show unchanged
completion time for the project, but in fact all the non-critical
jobs may be slipping badly and float is being consumed.
Problems with the computing system

One of the major reasons for CPMIPERT falling into
disfavour is the inadequacy of the computing system used in
the past, and the overhead cost of specialist planners and
data processingpersonnel to handle this side ofplanning. In
practice, though there is discussion over the size of the
network that justifies the use of a computer, it could be
reasonably stated that 'if you don't need to use a computer
for CPMIPERT, you don't need to use CPMIPERT'.
In the 1960s, computer-based CPMIPERT planning
systems used batch computers implying a complex and
powerful sophisticated system, designed to do specific calculations at discrete intewals. Thus the CPM/PERT programmes were also batch-oriented and were not, and could
not be effective project management information systems.
These systems were able to do the calculations involved in
148
CPMIPERT analysis quite satisfactorily, but suffered from

many of the disadvantages of batch computing. Batch
computers generally require data processing staff to operate
them, who act both as an interface and a barrier to access to
the computing system by the project staff. Specialist staff
were required because of the complexities involved in using
these machines; involving such things as job control language, use of punched cards, file handling and formattingof
data. This inevitably meant delay in getting a turnround of
the plan, no real quick access to data, and thus no 'feel' or
interaction with the planning system. In addition, runs of
the project planning computer program had to queue with
others jobs waiting to be done by the company's computer,
and thus response tended to be slow. The best that could
probably be achieved was a 24-hour turnround of the
program, though in some cases this could take a week. This
was completely inadequate, particularly in the early stages
of plan development. Computer cards had to be punched
and verified in a fixed format to hold the base data for the
plan. Inevitably there would be mistakes, such as data out of
place in the formatted card activities omitted, duplicated
activities, activities placed in a loop to give an infinite
completion time and activities with no connection, that is,
dangling activities. It would probably take several runs
before a satisfactory output would be obtained and this
could involve a considerable elapsed time.
Thus the speed of development of a plan was slow, just
when it was critical that it should be fast to guide the initial
work and identify critical activities andproblems. It is also at
this stage that it should be vossible to use a comvuter-based
planning system to simulate alternative ways of carrying out
the vroiect
- and for manpower planning. If this takes several
days to get the output from-each akernative, then it is
unlikely that this strategic use of planning will be carried
out Similarly, when changes and delays occur, there will be
delays before management can evaluate their effect and the
different courses of action arising from them, and this will not
be real time planning. It will also not be possible to query the
system to extract information; report formats will also be
A
inflexible and this will discourage its use as a management

information system.
Conclusion
Thus for a number of reasons network methods have not

been as successful as was once anticipated. Some of the
reasons for this were directly attributable to difficulties with
the technique, some to the state of colhputing technology in
the 1960s and 70s and some to general problems with the art
and science of planning large projects. However, management experience, education and general professionalism in
the art and science of planning has increased, and many
more managers and engineers are using plans as tools to
assist them, on a real time basis, in managing their projects.
In practice this is probably more critical to the success of
planning than the actual planning technique used The
essential factor in project planning is that all the work
should be formally scheduled in a mannerwhich will permit
the evaluation of actual progress against plan andwhich will
identify the interdependencies of individual activities, and
not which technique is used.
In many small and medium-sized projects, bar charts will
be the favoured primary planning technique, but whenever
it is necessary CPMIPERT should be used to support them.
In larger projects, although hierarchical bar charts can be
used actually to plan and control the project, CPMIPERT
should also be used to handle interrelationships. In addition
computer-based CPMIPERT is a very efficient way of
producing and updating bar charts, and for manpower
planning. In fact its use can almost be completely justified
for its ease of handling barcharts. WheneverCPM/PERT is
the primary planning technique, it should always be complimented by the use of bar charts. In many cases centralised
planning should be carried out using both CPMIPERT and
bar charts, and then bar charts issued to functional managers
and supervisors, who often need not be aware of the use of
CPMIPERT. CPMIPERT and bar charts both have advantages and disadvantages, but they are both effective planning
150
techniques which should be used in conjunction with each

other, and with other techniques such as manpower histograms, S curves and milestone charts. The combined use of
these techniques is now much easierwith modern computer
systems and project planning program packages.
Modern computer-based techniques
Where network methods have a major advantage is that they

are amenable to com~utermani~ulationand can be
integrated with corn6ter-based Iproject management
information systems. In this computer age it is illogical not
to use a computer ifit can be ofassistance to management. In
the past computer-based project systems have had a bad
reputation in project work, primarily because they were
based on batch computers, operated by data processing
personnel and the packages, which had been devised by
data processing or operational research staff, were sometimes not very suitable for practical use.
Today, revised project planning computer program packages and modem computing systems can overcome the
problems with project planning outlined in this chapter.
From the mid-1970s revised. or new vroiect management,
planning and control computer packages, based on practical
experience and the identified needs of users, began to be
used with interactive computers. These systems are capable
of being effective project management planning, control
and information systems and are now rapidly coming into
use. They allow the project staff direct access to the computer,
with more or less immediate response to the requests for.
information, construction or updating of the plan.
These interactive project management systems were
generally based on research into just what users wanted
from such a system, and notwhat data processing specialists
thought should go into it. Thus in general they have the
following advantages over the older systems, namely:
1 They are easier to use, both for the novice and
experienced user.
151
2 They are much faster in response.

3 The manager, planner or engineer can use them
directly with no data processing interface to go
through.
4 They are very flexible in that they can produce bar
charts, manpower plans, float charts and S curves if
required for any part of the project
5 They are more powerful than the batch systems, not
necessarily in the straightforward calculations, but
in the sorting and manipulation of data and output.
6 They are more tailored to the practical needs of the
project staff.
7 They can be integrated with data base management
information systems.
Modem project planning systems
The modern project planning packages inuse today arelittle
different from older packages in the basic calculation
routines, but are very different in the framework in which
they are used. These packages should have the following
characteristics.
Ease of use
The project planning program and the computing system it

runs on must appear to the non-data processing-oriented
project user to be simple and easy to use. It should be
interactive, not only for the reasons outlined previously, but
also to achieve acceptance of the system by having contact
with it and to achieve the man-machine combination, which
can be more flexible and powerful than either working on
their own. The system should be based on an analysis ofuser
needs and desires, not only from a general project management point of view but also for the managers involved on
any individual project. Thus it should be able to be used for
simple planning, or for sophisticated planning; it should be
able to be used for CPMIPERT and bar charts used jointly;
for bar charts as the primary technique or for CPMIPERT as
152
the primary technique. User instruction manuals should be

as simple as possible with a kindergarten level introduction
to planning the simpler project. The program should have
the facility to be used at beginner level, with a 'help' query
system and detailed prompts, and at an expert level without
these user aids. It should also give a fast response.
Project planning program
The program should be just as much oriented to bar charts,

manpower planning, S charts and milestones as it is to
CPMIPERT. It should be a 'project planning' program
package and not just a CPMIPERT program.
Integrated project management information system
Many CPMIPERT programs have PERTICOST facilities

where an estimated cost, actualcosts and avariance analysis
can be used for each activity. A more modem approach is to
link the data files for the project planning system to a data
base management system, so that not only can this method
of analysis be carried out but also so that a more modern
performance analysis system can be implemented based on
the use of the work breakdown structure, cost accounts and
work packages. This provides a basis for an integrated
project management information system.
Input facilities
This must be simple, fast and as far as is possible limit the

chance of error. Either free format should be used, or
formatted screens on visual display units, with prompts for
beginners should be available. It must be easy to set-up, use,
copy and edit multiple data files to hold the data on
alternative plans. The bulk of the data may be input-using
punched cards, at an interactive terminal or held in a data
file, and a flexible interactive data editor must be available to
correct and add to the data. Library files must be available to
hold standard planning modules, described later, which
only need to be edited to add to this project data base. It must
153
be possible quickly to establish and modify alternate calendars to provide the time base for the plan.
There must be comprehensive error checking routines
which first of all validate the raw data to see if it is complete
and within the bounds of reason, and secondly check the
logic of the arrow diagram for errors, such as loops or
dangling activities.
Operation
The basic calculation routines are well established for both

activity on the arrow and activity on the node, but it should
be possible on the same data base to analyse all levels of plan
used, by summarisation and skeletonisation. In addition to
the straightforward time analysis, the program should be
able to do manpower planning on four bases:
1 Time limited.
2 Resource limited
3 User priority code.
4 Manual interaction.
The program should be able quickly to simulate different
logic, activity times and resource availabilities in the plan
development stage and when replanning the project, using
X schedule
alternative data files. It should be bossible to ~ Ithe
for any activities described, that i t establish scheduled starts
(SS) and scheduled finishes (SF). On updating runs it
should be possible quickly to input actual starts (AS) and
and percent complete estimates for any
actual finishes (AF)
activity.
Output
The program should have a flexible report generator, which

on request should be able to generate the following output
for the project as a whole, for the hierarchical planning
system, for the work breakdown structure elements, organisations, cost accounts and work packages and sort for any
factor or combination of factors:
154
1
2
3
4
5
6
Traditional time analysis.

Bar charts.
Manpower histograms and tables.
Milestone charts.
S charts.
Variance analysis from plan, and from update to
update.
There should also be a connection to the data base

management system for cost and schedule performance
analysis.
Modem project planning methods

The project manager and planner can overcome many ofthe
problems described and effectively plan and control their
projects, with the use of modern computer-based project
planning systems.
These systems are much more effective in achieving good
communication than the older batch systems because they
are designed for, and can be used as real time project
management information systems. The project staff can
through a computer terminal extract information more or
less immediately on request or on a routine basis. This
information can not only include normal time and resources
data but also many other items of information. This is done
by extending the data records held in file for each activity
a n d or, interlinking the data network data files on a data
base management information system. For example, such a
computer system can hold for each activity the information
shown in Figure 5.11.
The standard set of reports produced by such a system
consists of:
1 A time analysis.
2 Bar chart.
3 Resource histograms and tables of manpower
requirements.
This can be produced for the project as a whole, as is shown
Description
Whose r e s p o n s i b i l i t y
Preceding event number
Succeeding event number
Duration
Number of r e s o u r c e s
For each r e source:
Name
Amount p e r time period
T o t a l amount e s t i m a t e d
T o t a l amount a c t u a l
ES
Earliest start
LS
Latest s t a r t
EF
Earliest finish
LF
Latest f i n i s h
SS
Scheduled s t a r t
SF
Scheduled f i n i s h
TF
Totalfloat
FF
Free f l o a t
SF
Scheduled f l o a t
AS
Actual s t a r t
AF
Actual f i n i s h
% complete
FF
Forecast f i n i s h
EL
Budgeted c o s t labour
EX3
Budgeted c o s t m a t e r i a l
ACL
Actual c o s t labour
ACM
Actual c o s t m a t e r i a l
I
F
User f i e l d s ( t o add any e x t r a s o r t codes r e q u i r e d ,

d e s c r i p t i o n s , o r any o t h e r information)
Figure 5.11 Data base for an activity
156
in Figures 5.10,5.12 and 5.13, for the individual department

or organisation, for each profession or trade, or any other
factor as desired, for example, one area of construction site,
one unit of a chemical plant. Figure 5.14 shows a standard
set of reports for the electrical-instruments system design
function in the project used for illustration. In addition S
curves and milestone reports can be produced as desired for
the analysis and control of progress.
The ability to sort output is a very powerful and extremely
useful facility and means that reports can be produced that
are sorted by responsibility, by date, by resource, by criticality, by function and by many other factors. For example,
reports for the following can be produced in any form
desired, and these factors can be combined together for
special reports on request:
1 All activities due to start next month.
2 All activities in a particular area.
3 All activities which are the specific responsibility of
an individual, department or company.
4 All activities using a specific resource.
5 All activities with zero float
6 All activities with free float.
7 All activities with less than ten units of float
8 Number of fitters required over the next sixmonths.
9 The activities these fitters would be working on in
order of total float.
10 All jobs whose float this month is less than float last
month.
Thus, for example, a supervisor or manager could be given a
time analysis, bar chart and resource requirements table or
histogram for all the activities for which he is responsible in
the next three months. Such a system is not only effective for
communication but is invaluable as a source of information
for management.
Using these systems several methods can be used to
monitor, analyse and display progress. The most common
are:
1 Bar charts.
2 Time analysis with progress reports.
157
3 Manpower curves.
4 S charts.
5 Milestones.
It is relatively simple to produce bar charts showing both

planned and actual work on activities by computer for the
project as a whole, or sorted by area, responsibility, current
jobs, or for activities which are not on schedule.
Time analysis reports can also be produced showing
scheduled and actual start and finish for each activity,
plus reports on percentage complete and forecast completion
dates, again sorted in any manner desired, as shown in
Figure 5.15.
Manpower graphs can also be produced by the machine
showing planned manpower build up and actual manpower
build up as shown in Figure 5.13. They can also be produced
for total manpower, and for individual trades.
S curves
The computer-based system can also produce that very

effective control tool. the S curve. An S curve is a mauh of the
cumulativevalue of manhours, percentage complete, orcost,
against time. The vertical scale can be manhours or cost or
tce percentage of planned manhours or budgeted cost, that
is, percentage complete. This graph generally takes the form
of an S because most projects have a slow start, followed by a
longer period of relatively constant activity at a higher rate of
activity, and finally a falling off of this rate of activity to give
a slow finish.
The S curve is a very sensitive tool for the analysis and
control of progress, whether it be based on manhours or cost.
In the control of cost, their use has become very sophisticated,
and is described in more detail in a later chapter. The big
advantage of S curves is that they can be used to identify
trends at an early stage, because they can monitor both the
rate of progress, and also the acceleration or deceleration of
this rate. In manpower analysis the slope of the curve
represents the rate of expenditure of manhours, that is the
-.
- - - - -----
BAR CHART
29
ACTIVITY
MEN I
10
20
30
40
50
60
------------------------------+---------+---------+---------+---------+---------+---------+---------+
1
2
3
4
5
6
7
A
9
10
11
12
13
14
15
lb
17
18
19
20
21
22
25
26
27
28
BASIC DESIGN
MECHANICAL DESIGN 1
ELECTRIClL DESIGN 1
PURCHASE MECH.PbRTS
MECHANICAL D E S I G N 2
S P E C I F Y ELECT.MOTORS
PURCHASE ELECT.PIRTS
E L E C T R I C A L DESIGN 2
INSTRUMENT DESIGN
PURCHASE INSTRUMENTS
D E L I V E R MECH.PARTS
MECH.DRAUIN6 OFF.
PURCHASE EL.MOTORS
D E L I V E R ELECT.PRRTS
ELEC.DRIUING OFF.
1NST.DRIYINGOFF.
D E L I V E R INSTRUMENTS
F A B R I C A T E MECH.PARTS
D E L I V E R MOTORS
F A B R I C A T E ELCT.PARTS
ASSEMBLE MECH.1
ASSEMBLE ELEC.l
ASSEMBLE MECH.2
RSSEMBLE ELEC.2
F I N A L ASSEMBLY1
F I N A L ASSEMBLY2
16
12
16
4
12
4
IXXXXXXXX
I
xxxx---------1
XXXX
I
I
XXXX- - - XX- XXXXXXXX
!
I
XXXX- -
lb
12
4
I
I
lb
4
0
20
16
0
70
0
60
40
36
40
3b
24
24
:
I
I
I
I
I
I
I
I
1
I
;
xx-----------xxxxxx----------
xxxxxx~x------------
xxxxxxxx------~---~xwxxxx----------
XXXX- - - XXXXXXXXXXXXXXXX- XXXXXXXXXXXX
xxxxxxxx---------~--
XXXXXXXXXXXXXXXXXXXXXX- -
x~xxxxxx----------
XXXXXXXXXXXXXXXXXX- - - XXXXXXXX
XXXX- - - XXXXXX
XXXXXXXXXXXX- - - XXXXXXXXXXXX
XXXXXXXXXXXX
XXXXXX
------------------------------+---------+---------+---------+---------+---------+---------+---------+---------+
Figure 5.12 Computer-produced bar chart (showing activity time and total float)
70
Figure 5.13
Manpower chart
159
n r r I.
7%
m r wmrn or Bar *rromrr MI T
IIRIII~
II
Figure 5.14 Plan for electrical instrument design

160
velocity, and the rate of change of the curve represents the

rate of build up or run down of the momentum of work on
the project, that is, the acceleration or deceleration of the
pace of work. S curves can be used to represent the project as
a whole, work-breakdown structure items, organisational or
functional work packages. Figures 2.2 and 5.16 show the
planned percentage manhours S curves for work on the
design and construction, respectively, of a petrochemical
project
Three elements of data are required to be graphed on an S
curve for effective control of manhours. These are:
1 The planned cumulative expenditure of manhours
against time.
2 The acutal cumulative expenditure of manhours
against time.
3 The cumulative manhours equivalent of actual work
completed, that is, earned value.
The slope of the curve at the start represents the initial rate of
work on the project. If the gradient of the actual manhours
curve is less than planned, then it is obvious that there is too
slow a start. Thereafter the first critical point is where the
curve should turn up at the bottom of the S. If this critical
acceleration of work on the project does not occur as
planned, that is, the S curve does not turn upwards, the
project is going to be delayed, no matter what is done. Once
the curve does turn up, the slope of the curve shows the rate
of progress. If work is not progressing as fast as planned,
the slope of the actual curve very quickly and obviously
becomes less than that of the planned work curve. At the top
of the curve if the work does not decelerate as planned then
there is going to be an overshoot. This is because it is
generally not possible to finish off a project without the
slowing up of the pace of work, as is represented by the top of
the S.
For example, Figure 5.17 shows the S curves for the
electrical instrument systems design functions for the functional plan shown in Figure 5.14 at week 17 of the project
The planned manweeks expended curve is simply the
cumulative version of the manpower graph shown in Figure
A c t i v i t y Description
Basic design
10
10
20
Specify e l e c t r i c a l motors 20
35
20
45
Mechanical drawing o f f i c e 30
60
Instrument design
20
50
E l e c t r i c a l drawing o f f i c e 45
70
Instrument drawing o f f i c e 50
80
SS
SF
AS
AF
Complete
Figure 5.15 Time analysis for electrical design and total drawing office
Forecast
Completion
Men
Figure 5.17 S curve for electrical design
5.4. The actual cumulative manweeks expended is based on

the project records. Both these curves are fairly straightforward to obtain data for and to construct, but can be very
misleading without the third curve. that is. the earnedvalue
of manwegks expended There is often difficulty in obtaining
reliable estimates of this cumulative manweeks eauivalent
of actual work completed and this is one of the reasons for
the introduction of the work package concept, which can
give an objective estimate of earnedvalue. If a work package
is completed, the earnedvalue is the manweeks estimated in
the plan and budget for its completion, despite the fact that it
may have actually taken considerably more manweeks to
complete i t
The method of analysis can be observed in Figure 5.17 at
three points, namely, weeks 10, 15 and 17.
Week 10
The activity labelled basic design is just complete and 14
men were employed on it for 10 weeks, instead of 16 men for
eight weeks as planned The S curve clearly shows this
deviation from plan, together with a poorer than estimated
performance and can be calculated as follows
(a) Planned cumulative manweeks= 160 ie., 16 x 10
(b) Actual cumulative manweeks = 140 i.e., 14 x 10
(c) Earned value
= 128
i.e., 16 x 8
In this case the earned value is the planned manweeks to
complete basic design. There is little need to use these
calculations to demonstrate the deviation from plan and
completed performance, as the differences in the slope of
the three curves clearly show these factors. The horizontal
difference between the earned curve and the'planned' curve
also shows any deviation from schedule; in this case the
project is already two weeks behind schedule.
Week I5
The first stage of electrical design is now complete and 16
166
men were employed, as planned, although this activity took

one week longer than planned. Thus the slope of the 'actual'
curve is the same as the 'planned' curve for that stage of the
work, but the 'earned' curve's slope is less than that of the
actual. This clearly identifies performanceless than planned
and brings it to the attention of the project manager for
further investigation
The project is now three weeks late, as shown by the
horizontal difference between curves and performance can
be calculated as follows
(a) Planned cumulative manweeks= 290
(b) Actual cumulative manweeks = 220
(c) Earned value
= 192
(that is, the estimated manweeks to complete basic design
and electrical design I).
It is also obvious at a glance that the pace of work has not
accelerated as planned and that the deviation between the
curves is increasing.
Week 17
Although the slopes of the actual and earned value curves
have increased, they are still less than that of the planned
curve. The project is now approximately four weeks behind
schedule and an extrapolation of present performance can
now be made to forecast the completion date of this segment
of the project and the number of manweeks that will be
required. If performance is unchanged it can be estimated
from the curves that this segment of work will take 26 weeks
instead of 20 as planned, and 505 manweeks instead of 432.
The results to date can be calculated as follows
(a) Planned cumulative manweeks = 350
(b) Actual cumulative manweeks = 275
(c) Earned value
= 235
Earned value
Manpower performance index = Actual
value
235
275
Estimated total manweeks

Planned manweeks
Manpower performance index
167
-432
0.85
= 505
It is obvious that not only are the promised number of

men not being committed to the work but that these men are
not achieving the expected performance.
Although the points demonstrated on this segment of a
project are relatively self-evident from other factors, this is
not always the case for large segments of work and the
project as a whole. The S curve being a cumulative curve is
sensitive to deviations which although small in themselves
in any one period, can build up to a significant deviation.
When the data for all the activities on a project are
consolidated, S curve analysis gives a concise and clear
picture of performance on a project. Figure 5.18 shows the
complete project, the planned curve being the cumulative
version of Figure 5.13, and the same form of analysis can be
carried out For example, at week 32
(a) Planned cumulative manweeks = 1,575
(b) Actual cumulative manweeks = 1,250
= 1,025
(c) Earned value
The project is approximately 6 weeks behind schedule
because in the 24 weeks, fewer men were committed to the
work than planned and they did not achieve the estimated
performance. The acceleration of work into the main phase
of the project was 4 weeks behind schedule and though the
planned number of men are now being committed to the
work, they are still not achieving the required performance.
The manpower performance index can be calculated on
three bases, that is, on the project to date, on the main phase
of the project, that is, week 28 to week 32, and for individual
segments as described above. For example
Manpower performance index on
project to date
= 1,025/1,250 = 0.82
Manpower performance index on - 1,025 - 750 = 0.85
main phase
1,250 - 875
If the earned value curve is extrapolated it forecasts the the

project completion will be 13 weeks late. If the manpower is
unchanged it forecasts that either 4,604 or 4,442 manweeks
will be required to complete the project instead of the
planned 3,776 manweeks.
Thus these S curves and the information they represent
can be used at any time to determine how much work is
actually completed, the efficiency of working the rate of
working and to forecast the time to completion and the
actual manpower required.
One of the skills a project manager must have is the ability
to recognise trends, or deviations from plan or budget, at an
early stage. Everyone can recognise a trend or deviation
when it is well established, but by that time in most cases it is
too late to do anything about it. To be able actually to
influence the success of the project, rather than be carried
along by its momentum, a project manager must be able to
recognise a trend at an early enough stage to be able to do
something about it. This may be partly an intuitive skill, but
this can be enhanced by the intelligent use of the analytical
techniques available, of which the S chart is one of the
simplest, but one of the most useful.
Milestone charts
Milestone charts are another useful and simple planning
and control tool which can be used as follows:
1 As an effect& control tool.

2 To relate one update of a plan to others.
3 By itself, as a simple form of planning and control,
for example, in procurement.
4 As a useful summary plan.
5 To show trends in the slippage of completion dates
for parts of the project.
Milestones are specific and obvious points of achievement
marking the start or completion of stages of a project. These
can be major milestones identifying major phases of a
project and minor milestones for intermediate points. They
170
may be, but are not necessarily confined to nodes or events

in the critical path or arrow diagram. The estimated completion time of each milestone is determined from the initial
version of the plan. Thereafter the re-estimated, or actual
completion time, is determined at each update of the plan.
This information can be presented in tabular form as shown
in Figure 5.19 or a milestone chart can be constructed as
shown in Figure 5.20.
In this chart the estimated date of each milestone at the
initial construction of the plan is plotted along a horizontal
time scale at the top of the chart. Thereafter at each update
of the plan, the actual or re-estimated date of each milestone
is plotted as shown. As they are plotted, each point is joined
up to the preceding one to produce a series of curves, each of
which shows the trend in date for that milestone. There
may be an optimistic tendency to straighten out these lines
at the right hand side. hoping to make up time on these items.
Generally, however, delays in the early stages of the plan are
likely to be reflected all the way through, as illustrated, and if
this seems realistic, it is better to make an early estimate of
slippage in project completion date, than be faced with a
major revision nearer the end.
This chart can show: the estimated completion date,
changes in estimated completion dates between updates,
trends towards changes in these dates, causes of these trends,
progress, if any, towards returning to schedule, and the
actual date of the milestone to give a historical record. It has
the advantages that it can permit any significant point to be
followed, it can show overlapping items and it is simple and
easy to read.
Management of float
Many projects become late because of slippage ofthe critical

path, but many also become late because of slippage of 'so
called' non-critical jobs. It is essential that the float, that is,
spare time on non-critical jobs be monitored from one
update of the network to another, to check for slippages. A
general picture can be obtained by the printing out of a
171
graph, or histogram showing the number of jobs with

different amounts of float at each update. This, however, only
produces a general picture which is not too clear. An
alternative method is to print out on a time analysis the
previous, or initial float and the present float. Alternatively,
the computer can be instructed to print out this information,
where present float is less than previous, initial, or a
specified level of float. This, however, may result in large
amounts of print out, which will not be looked at. An
effective way of monitoring float consumption on a shorthand, exception basis is to monitor the free float on a path.
This can also help to convey the knowledge that float
belongs to a path through the network and not to any
individual jobs.
Except on the critical path, all subnetworks are joined to
the master network by free float. This path float is shown as
total float on activities along the path and as the free float on
the last activity on a path before a junction event. Figure 5.21
shows a skeleton bar chart, with consolidated paths and
their free float.
By monitoring the activities with free float a check can be
kept on the consumption of all float in the network. This can
be done for all paths in a network, if there is a limited
number as in Figure 5.22, or it can be done for only those
paths that show a decrease in free float The project staff can
then follow up a particular path by simply requesting a print
out of data for all jobs on that path to find out where the
slippage is, using a path trace option or sorted output. Thus
modern computer-based systems can effectively monitor
slippage on non-critical paths and changes from one update
to another, on a total or an exception basis and can be used
to investigate these further by using the project management'
information system.
Operational plan and schedule
If the basic critical path method is used to &an and control a

project, all that is produced is a permissive framework of
start and finish times for the activities. Who does what,
Event time (Calendar week)

Original
Schedule
complete
12
E l e c t r i c a l items
o r d e r s placed
complete
20
Mechanical items
delivered
24
Start fabricating
electrical
32
S t a r t assembling
mechanical
38
S t a r t assembling
electrical
40
S t a r t f i n a l assembly I
46
S t a r t f i n a l assembly I1
58
P r o j e c t complete
64
Figure 5.19 Milestone report week 48
Review 1
Week 12
Review 2
Week 24
Review 3
Week 36
Present
Forecast
Actual
Figure 5.20 Milestone chart
173
174
when, is then decided in an ad hoc manner by junior

management and supervision. This almost always leads to
an early consumption of the float available on non-critical
activities, resources imbalance and slippages in the completion time of the project. Someone must decide who does
what, when, but it must be done in a planned manner,
integrating resource scheduling and time scheduling of
activities. It may still be necessary and advisable to leave
day-to-day management, or even week-to-week management
of the work to junior management and supervision, but the
amount of latitude delegated to them, must be limited. This
day-to-day, week-to-week work, must be planned and carried
out within the framework of a plan, but this framework must
bZ tighter than the conventional earliest and latest times for
an activity. What is required is thescheduledstarz and$nish of
activities and the resources planned. Ideally this detailed
planning of the project work should be done in conjunction
with iunior management and suoervision.
~ G - t o - d amanagement
~
ma; require changes to the
schedule due to a hundred and one factors which cannot be
anticipated, but variations must be kept within a tight
framework to give management a chance of keeping to
schedule. A scheduled start and finish date for activities
does not mean the simple suppression of float, but the
establishment of an agreed start and finish for activities by
project management, planning staff and junior management
and supervision, who then attempt to control the work to this
schedule. This has the big advantage of being 'their' schedule
and thus they will more likely be committed to it and it is
more likely to be achieved.
Thus in planning the project, the preparation of a critical
path diagram and its time analysis are merely the start of
planning. The next step is to determine the resources
required by this plan. This is done by the computer in a
similar manner to how it is done manually with bar charts.
The data base for network analysis contains the manpower
and other resources required (for example, cranes) and the
computer can aggregate them in a similar manner to a bar
chart for a plan based on both the earliest and latest start
times of activities. In addition, non-critical jobs can be easily
'2
176
scheduled to start at some percentage way along the float

This gives the initial graphs and tables of the manpower
required, both for individual trades and for the total manpower.
This pennits the manager or planner, to examine the
manpower profile and compare it to the desired build up
and run down of labour, and the labour availability.Where a
smooth build up or run down is required, orwhere resources
exceed those available or desired, perhaps to avoid congestion, activities have to be adjusted hopefully within their
float, to endeavour to achieve the desired manpower plan, as
with bar charts.
There are at least three ways of doing this with computerbased project planning, namely:
1 An interactive semi-manual system.

2 An automatic resource levelling system.
3 A combination of the above.
In the interactive method, manpower profiles for the total
project and for individual key trades are printed out and
examined. When the total project manpower profile is
undesirable or where the key trades profile is fluctuating
widely or exceeds the numbers available, 8 time analysis and
bar chart are printed out. It is also possible automatically to
print these activities contributing to peak demands for any
particular trade. Then by inspection,that is, manual planning,
activities can be given scheduled starts andlor resource and
time requirements varied to achieve an acceptable manpower profile.
To illustrate the flexibility of this combination of human
judgement, machine interaction and the quick evaluation of
possible alternatives, consider the following example. Figure
5.2 shows the manpower charts for electricaVinstrument
designers, and for the drawing office. In both cases the
manpower assigned to this project varies erratically and
probably at the peaks exceeds that available. Good manpower planning would endeavour to avoid these peaks, and
have a nearly constant allocation of men to this project.
Figure 5.14 shows the full sorted output for the electrical/
instrument design activity and shows that two activities have
178
float and thus could be rescheduled without delaying the

overall project. By inspection of the bar chart it can be seen
that if instrument design can be delayed by eight weeks, this
would smooth out the main peak and yet the activity would
still have four periods of float. To do this simply involves the
following answers to prompts from the program package:
'DO YOU WANT TO RESCHEDULE ANYACTIMTIES?

? YES
'WHICH A C T I V I m
? NSTRBES.
'WHAT IS THE NEW SCHEDULED START WEEKNO?'
? 20
The program will then recalculate the network data and
produce any reports desired. Figure 5.23 shows the new
manpower chart for the resource in question. This gives a
reasonably satisfactory manpower plan and the next stage is
to see what effect this has on the other critical activity, that is,
the drawing office as shown in Figure 5.24. In fact it has
reduced the peak drawing office manpower requirements
from 52 to 36 but the profile is still too lumpy. By repeating
the process with the instrument drawing office we can get the
manpower chart shown in Figure 5.25. This is still not too
satisfactory and has in fact used up all the float on that path.
A more reasonable plan would be to extend the work on the
mechanical drawing office to take 10 weeks instead of 6,
employing a maximum peak of 28 men on the project and
start work with 8 men on the instrument drawing office as
soon as the mechanical drawings are complete. This would
reduce the peak manpower requirements and still retain two
weeks' total float on the instrument path. Thereafter a full
report would be produced to examine the overall effects and
sorted reports for each activity to check the effects on
individual resources.
However, with large jobs this can get difficult, time
consuming and require many attempts to achieve an
acceptable manpower profile. In addition, as the job
progresses and changes occur, updating the manpower plan
can bevery tedious and time consuming. Modern computerbased systems allow the maximum number of tradesmen
per period to be specified and the computer automatically to

endeavour to carry out this scheduling, using a scheduling
algorithm.
There are two types of scheduling carried out on modem
computer-based systems, namely, time-limited and resourcelimited scheduling. Scheduling starts from the earliest start
schedule. It takes account of the resources required for each
activity and their availability, delaying activities where
necessary until the required resources are available. Ideally,
this can be achieved without missing any target completion
dates or delaying the completion of the project as a whole by
rescheduling activities within the float available. It may not
be possible, however, to satisfy the constraints of target dates
and resource availabilities simultaneously. I n this case
either the target dates must slip or more resources must be
provided. The two types of scheduling, time-limited and
resource-limited, differ only in what happens if both the
time limits and resource limits cannot be satisfied simultaneously.
If it is the target dates that are all important, then timelimited scheduling will take account of the resource constraints and will attempt to satisfy them, but will provide a
schedule that meets the target dates. It will do so even though
the requirements for one o r more of the resources may
exceed the available level for some part of the time. Activities
which use resources over the normal availability level set for
these resources, will be tagged for future reference and
management attention.
If the resource constraints areunchangeable, then resourcelimited scheduling will provide a schedule which restricts
the total resource requirements to the available levels, even
though this may result in missed targets or a delay in the
project completion date.
Systems using scheduling algorithms generally require a
larger and faster computer than others and this process can
take a considerable amount of time for larger project plans.
The human touch tends to be lost and it is difficult for
automatic scheduling to take into account complex constraints and intangible factors. However, both methods can
be combined to give some human interaction and some
.-a2
u
!?i
* I *
-- .-..
2 2"
---
he-
-----
NNNNN
F"-"-,--
I
.
, r, %
r., N
1I.I
.*CillS
.............
.CTIV,II
I
30
T
6
12
>I
"tc".la"ulnr
orr.
(1
I,
CtIC.DI1V!lb
,,9,.Dt"","'
OFF.
TI*
I" T I CODf """I<" Or I l t 0P110" D I E l l t l i e F01 ,YE IlDlS.,,,
- 1
TlPl
I#
THC
ills
LO
71
111
43
10
OFF.
LI
LI
GF
Is
10
11
211
:1
>1
I*
LT
I4
3:
12
11
to
DT 1C 1111c.1
..
rr
rr
I0
,V",l.,ll,
RY11E1 OF 1HE I I I O Y I C T Ic?O.1 I I D U r I I I
'1
IU C I Y R ,
.........
3 ICT~VIVV
111
II
0 "Ct".DR"Y,"I011.
8.
I,
c,tr.II*YI*L
OFF.
I9
(4
I",T.PI*"I*I
PFI.
I*
lo
<o
>a
re
ilxX""
xxXxIXXIxI"I
~x",xx*,---.
Figure 5.24 Drawing office plan

(after rescheduling of design plan)
181
70
a
4
C",.,".,
-- -- -- -- -- -0 0 0 0 0 0
NNNNNN
automatic resource planning to handle large and complex

jobs.
With such systems project and construction managers can
use simulation to help them evaluate alternative decisions
and changes that occur. For example, the user can enter any
changes he requires to activity durations. A rapid analysis
would then enable him to discover immediately how the
altered durations affect the summary network or how the
changes modify a specific area in the network Resource
simulation allows the user to raise or lower resource availability levels and to monitor the effect this has on the
completion time and on other resources. Thus such a
simulation gives a close approximation to the project
resource requirements that can be used later for a fully
detailed analysis.
Planning library modules

Many projects are completely new in concept and structure
to an individual company, but many others are relatively
similar and are characteristic to the industry or the company.
They may be widely different in detail, but he similar in
general structure; for example, a petrochemical project or an
aerospace weapon project When a company deals with a
stream of similar projects in such a situation, considerable
time and work can be saved in preparing a plan by the use of
standardised logic diagrams, or in other terms, library
modules.
These are standard plans, networks or bar charts, or
elements of them, which represent the way work is normally
carried out on certain activities, or on complete projects.
They essentially represent an experienced planner looking
at an old plan for a similar project, or parts of such a plan,
and using it as a guide or format for the plan of a new project,
but taken to a greater degree of sophistication When project
plans are examined for relatively similar projects it becomes
obvious that the general structure of their plans is naturally
itself similar. In engineering and construction an unreasonable amount of time and manhours are required for the
184
drawing and re-drawing of very similar looking engineering

and construction modules in a plan. There may be considerable difference in detail, for example, in a petrochemical
plant the number and size of vessels, towers and equipment
will be different,but the general logic is the same. In addition
when the individual parts of a plan are examined, similar
patterns emerge, for example, the design, specification,
purchase, delivery and erection of a vessel generates a
similar logic for every vessel in a project and in all projects of
a similar type. The interrelationships involved in the design
phase and the construction phase also form a standard
pattern, and it is possible to have standard logic diagrams
showing the design process, the construction process, and
the interrelationships between them and for various parts of
these processes, for example, in piping or electrical design.
In addition, major items of equipment, such as a distillation
tower, may have a complete standard plan for the tower
itself, its internals, the supporting structure, associated
vessels, pumps, instrumentation and piping. There may be
similar logic diagrams for up to a dozen or more different
types of items in a project, for example, vessels, pumps, heat
exchangers, compressors, buildings, etc. In addition, each
individual activity in the chain can also be expanded into a
standard mini network or bar chart, for example, the design
of an individual item or the purchasing and expediting of
that item.
Thus, it is possible to have three levels of standard library
modules:
1 A micro module for an individual item.
2 A mini module for design processes, or an aggregation
of items such as a tower and its associated eaui~ment.
3 A macro module for individual stages in a'cokplete
project, or the total project itself.
Figure 5.26 shows a macro module for the design of a
chemical pIant unit and Figures 5.27 and 5.28 show micro
modules for individual parts of it.
These library modules can be either handled manually or
by computer. In a manual system a drawing or drawings
would have the standard arrow diagram or bar chart at a
level 1,2 or 3 for the type of unit or project involved This

would have the key events with event numbers widely
spaced. There would also be standard drawings for the
micro and mini modules for items going into this macro
module. A planner would then simply take out the number
of modules required for the individual project, add the key
interrelationship event numbers from the macro modules,
perhaps modify it slightly to tailor it for the individual
project, then number the events in the various modules and
add the names and other details.
With a computer-based system, the planning can be more
automated and instead of drawings, computer files would
hold the necessary details. A project plan for a petrochemical
plant having, for example, two catalytic crackers, four
boilers, twelve towers, fifty vessels, seventy heat exchangers,
four large compressors, six buildings, twelve tanks and 150
pumps could be assembled very quickly using this method.
The planner would input to the computer the list of these
items including names. unit numbers, a code as to the typeof
item and a code as to its size. The computer would hold the
standard logic diagrams for each item, together with the
details of the standard duration time and resource required
for each activity and for the various sizes of items. The
planner would input key event numbers, major interrelation
points and the computer would generate a plan and a data
base for the total project This would be reviewed, activities
added or deleted and data edited interactively by the
planner, before generating a final plan. There is even a move
to have a computer actually draw the resulting arrow
diagram using a graph-plotter. Thus, using this method,
either manually or by computer, a plan can be quickly
generated at the critical early stages of a project to guide and
control the work on it.
Consider a medium to large project, depending on one's
point of view, having anything from 500 to 5,000 activities on
a plan. Figure 5.29 shows a work breakdown structure for
such a petrochemical plant where the overall plan has 1,600
activities at a level somewhere between level 2 and level 3. It
has five units, four of which are similar in structure, only
one of which is exploded in any detail. The planning for
such a project would probably initially be done for each

individual unit, in this case using the semi-standard macroplanning module for four out of the five units. This macro
module will be made up of several medium-sized library
modules each having several micro modules. Thus, it could
be auicklv
, assembled and tailored for the varticular unit
and give the vertical planning integration'for each unit,
within this case. av~roximatelv200 activities for each unit.
When each unit ievel 2 plah is complete, sorting would
product a horizontally integrated plan for each phase or
function, and manpower planning could be carried out for
each sharedresource and the plans adjusted forthe resources
available, any priorities and for phased completion if
desired. This would give a satisfactory level 2 plan and
adequate integration and manpower planning. It would not
perhaps give a plan adequate for control and for efficient
functional planning. A further level of planning would
probably be necessary, which would explode the 200 activity
unit plans into four to six hundred activities to give a level 3
plan for each work breakdown structure element. In addition, or alternatively, the individual functional plans for all
the units would be exploded into a level 3 functional plan
which would cover such areas as the piping and drawing
office work for all the units.
~L~
The line of balance technique
In addition to the techniques previously described, one

other basic planning technique is sometimes used in some
projects, that is, the line of balance. The line of balance
technique is a graphical method of scheduling that was
originally developed to improve the planning and reporting
of an ongoing production process. This technique was
developed by the US Navy in the early 1950s and it is
especially suitable for low-volume new product situations in
integrating research and design with small scale production.
In these types of vroiects, some of the effectiveness of
planning and control oithe entire project is often lost during
the transition phase from development to prototype produc-
'7
Project
Unit
1
Unit
2
Unit
Unit
Off-sites
I
Flare
Cooling
Water
Boiler
Storage
tanks
1
Area
Layout
Vessels 8
Equipment
Electrical
Instruments
Foundations
Structural
Figure 5.29 WBS for a multi-unit project

(Each unit and off-site element can be expanded as is Unit 4)
Piping
Painting II
Insulation
tion and full production. This technique can handle this

critical transition stage. Its principal applications in project
work are:
1 Where the project is made up of a number of
identical units manufactured or constructed in
sequence, such as houses.
2 In a similar situation where the project consists of
first the design and then the production of a limited
number of units based on that design, for example, a
pre-production run.
3 As one reporting tool in one-off projects to give a
concise presentation of progress against plan.
The line of balance technique is concerned with parts, subassemblies, or various stages ofconstmction which make up
the finished flow of units. The most important single result
of the line of balance technique(L0B) is the development of
a schedule for the completion of the component parts or
stages of construction, which make up the final product or
project. Line of balance can be used to set up the initial
schedule for a series of units and it can be used to monitor
the progress during the production or construction cycle. In
a multi-unit situation such as a housing estate, it is difficult
to establish how work is progressing against plan, because
the various units will be at various stages of completion at
any one time. Some will be being completed, some will be at
various stages of partial completion and others will just be
starting. Similarly with limited production runs at the end of
a design and manufacturing project, the transfer of such a
project from design to production is always subject to poor
communication and loss of momentum. The line of balance
technique can provide a simple and easily understood
method of establishinga plan and controllingsuch situations.
All that is required in the line of balance technique is that
an objective made up of the number of units to be completed
in each time period is established and a simple plan made
for the production, or construction, of one unit. This first
requirement is normally termed the 'Objective' for the
production or construction process, where a quantity of the
end items is being made under contract, the objective is the
192
required delivery or completion schedule. The delivery

information used is of two kinds; the planned schedule
representing the contractual requirements and the actual
schedule showing the calendar date of delivery. The second
requirement is termed the 'Programme'. Once the objective
is determined, the next task is to define the programme,
which is simply the production or construction plan with
normally a bar or Gantt chart plotted against the lead-time
required before shipment. The monitoring of progress is
done against key operations or assembly points in the
manufacturing or construction cycle known as control
points. These are essentially events in a network analysis
system or the end of bars in the bar chart, or milestones in a
milestone chart.
Figure 5.30 shows a programme for a simple example, and
Figure 5.31 shows a n S chart from the cumulative delivery
schedule required and also the cumulative completion
schedule for these intermediate key events or milestones.
This represents the planned completion of the final units
and the various intermediate parts or stages. Then by
drawing a vertical line through this chart, it is possible to
determine how many units should have passed through all
the stages at that time. This is the line of balance and it
depicts the cumulative number of parts or units which must
have been completed by that date if the schedule is to be
maintained. Then by determining how many units have
passed through the key stages by that date it is possible to
measure progress against plan. This can be presented in a
compact form as in Figure 5.32 which shows the planned
completion at every stage, the actual completions at these
stages, and can also highlight any bottlenecks. For example,
the figures show that the first three stages are ahead of
schedule. whereas the remainder are behind schedule, with
the bottleneck being at stage 4.
Although the line of balance technique is more suitable
for small scale production or construction of similar units as
described, it is also used in engineering design and construction work as an ancillary analytical method of measuring
progress against plan and to give a concise picture of this.
Using the S charts shown in Figure 2.4 and 5.16, a line of
complete
sub
ass. A
complete
sub
ass. B
complete
sub
ass. D
complete
Q
M/C shop work
Events in L.O.B.
1.
Start work on A and B
2.
Sub-assembly A complete
3.
Sub-assembly B complete
4.
Sub-assembly D complete
5.
Machine shop work complete
Figure 5.30 Programme for a single unit in LOB example
Assemble & test
balance can be drawn at week 40 to give a concise pictorial

presentation of progress against plan shown in Figure 5.33.
This illustrates how clearly the line of balance method
demonstrates progress or the lack of progress against plan.
The line of balance technique therefore broadens the
other techniques available to allow the inclusion of repetitive
activities and is also a useful concise reporting tool. It aids
management analysis in that it highlights critical activities
that may delay the completion of the project and points to
problem areas, as well as to the present status of repetitive
activities.
Figure 5.33 LOB at week 40

(for typical petrochemical project)
6 The Planning and Control of the

Larger Project, or Portfolio of
Smaller Projects
The management, planning and control of the larger project,
or portfolio of smaller projects, is avery different'ball-game'
from handling the individual smaller project,
and is a very
- professional area of management and systems. The planning
and control techniques described in previous chapters are
perfectly adequate for the smaller project, but are merely
subsystems of the methodology necessary to cope with the
problems inherent in the planning and control of the larger
project.
It must be appreciated that where a firm, or a project
manager, is handling a number of small projects, i.e. a
portfolio of small projects, which, although independent in
every other way, use at least one common resource, then they
must be treated collectively as one largerproject Thus many
firms and managers are involved in managing the larger
project without being aware of it, and without the experienced
project managers and information systems appropriate to
the larger project. Therefore in the rest of this chapter, the
term larger project will be assumed to apply to both the
individual larger project and to a portfolio of smaller
projects.
The major factors leading to problems

Three major factors lead to the need for this other dimension
to handle the major problems involved in the planning,
organisation, control and integration of systems in the larger
project:
1 The large number of activities involved.

2 The large number of organisations and people
involved.
3 The lack of complete information to define all stages
of the project, which is a normal characteristic of the
larger project, i.e. change is endemic in the larger
project.
Construction of plans for the larger project
Defining and specifying the work involved in a large project

is a major task with hundreds, thousands and perhaps tens
of thousands of activities required to plan and control it
adequately. The ad hoc breaking down of the project,
described in the planning of the smaller project, cannot
attempt to handle the dimensionality of the problem and
more systematic methods are required to do this successfully
for the larger project. Another dimension has to be added to
the conventional planning and control methodology and
this other dimension is what defines advanced project
management.
Understanding the plan for a larger project
Unless planning is done very coarsely, with a consequent

loss of control, the amount of detail required to represent the
large number of activities involved in such projects results in
very large plans. Such large integrated plans are very
difficult to understand and to use. The span of comprehension of human brain varies from person to person, but in
most people it is overwhelmed by the dimensionalityinvolved
in very large project plans. It is very easy to become
confused and lost in such a plan and it is difficult to grasp in
one's mind the model of the project, which is all a plan is,
and actually manage it, rather than simply administer it.
In the case of the portfolio of smallerproiects
- - the plans for
the individual project are normally easily comprehensible,
but the interaction between proiects due to resource requirements is usually comple< &fficult to understand and
consequently often ignored.
200
Information and change
The above problem is compounded in the management of

the larger project in that this type of project is particularly
subject to change, due to the characteristic flow of information during the evolutionary life cycle of such projects. It is
more the norm with the larger project than the smaller one
that the information required to plan the later stages of the
project only becomes available as work on earlier stages is
completed. Despite this it is essential to integrate the
planning of all stages of the project into one integrated
whole.
Organisation of the larger project
In planning the organisation of the smaller project, the
global project organisation many only involve one or two
companies, and one or two departments or organisational
units in each of the companies. Even with this limited
number of organisations involved, there can often be problems with definition and allocation of responsibilities for
the individual activities or functions involved.
These problems are compounded exponentially with the
larger project, with often ten, twenty, a hundred or more
organisational units involved. Simply determining who
does, or is responsible for, what, can be a major problem and
omissions or duplications of effort are commonplace. In
addition s i m ~ l concevts
e
such as organisation charts cannot
represent adequately ;he c o m p l e x i ~of the responsibilities
and relationshi~sinvolved. Ad hoc methods of planning the
organisation a6d allocating responsibility canAonlyleah to
disorganisation and failure.
Integration of work plan and project organisation
The requirement is also not just to determine 'who does
what', but 'who does what, when'. Thus it is important to be
able to integrate the project organisation structure with the
project plan. In the smaller projects this can often be done
intuitively and it is simple enough to define. I n the larger
THE PLANNING AND CONTROL OF THE LARGER PROJECT
201
project plan. In the smaller project this can often be done

intuitively and it is simple enough to define. I n the larger
project this integration is more critical but much more
difficult and therefore systematic methods to do this integration of organisation structure with plan are required.
In the case of the portfolio of smaller projects, it is often
not appreciated just how many of these projects a functional
organisation is working o n with whom the functional
manager has to co-ordinate his work
Integration of the total project systems
The integ~ationof the project systems is a much greater
problem than simply the integration of the project organisation and plans. A critical factor in the management of the
larger project is the need for the integration of all the project
systems into one integrated project management information system, which should include:
1 The organisation structure.

2 The work plan.
3 The manpower plan.
4 The budget.
5 The control and information system.
6 The change control and data management system.
7 The materials management system.
I n the past, in many projects the financial control system
structure of cost centres could not be cross-referenced in a
logical and consistent manner with the work plan and its
activities. I n such cases, cost reports could not directly
reference reports on physical progress. In turn, neither the
financial system structure nor the work plan was integrated
with the management structure. Managers would have no
clear definition of what they were responsible for, when it
should be done, how much it should cost, when it was done
and how much it actually cost. It is essential for good project
management that none of these systems be considered in
isolation.
The integration of all these project subsystems into one
system on a common project structure with common cost,
202
information or management centres, is of tremendousvalue.

The lack of such integration ofproject systems has been one
of the reasons for the ineffective management, planning and
control of projects in the past
Control of the larger project
The problems in controlling a project described previously

have a much greater impact on the control of the larger
project and it is much more difficult to be 'in control of a
project of 3,000 or 30,000 activities, than one of 30 or 300
activities. Whereas control 'by involvement' overcomes
many of the faults of an organisation's control systems with
the smaller project, this is very difficult to achieve with the
larger project. Much more emphasis and importance has to
be given to the formal project control and information
system.
With the larger project a simple control and information
system. involving two levels of control centre, i.e. the project
and individual activity, can no longer be used. The control
and information system requires to be designed formally to
integrate and yet separate information for each organisation
involved and for each part of the project. Data needs to be
consolidated, filtered and analysed to give information to
avoid the common fault of data overload. The methods used
to analyse this data must also observe all the requirements
for a good control system described previously.
7 The Elements of Advanced

Project Management
The management of the larger project consolidates methods
and systems developed over the last twenty years. The basis
of this modem approach to the planning and control of
larger projects is the use ofwhat is called the work breakdown
structure. Although in the past projects were implicitly
structured in one way or another, as outlined in the introduction to planning, the concept of the work breakdown
structure (WBS) has crystallised a formal and systematic
way of defining and identifying what the component parts of
a .project are, breaking down the work to be planned,
structuring and integrating the project organisation and the
control and information systems. This approach to the
systematic organisation,
and conti01 of a project is
essential to the efficient management of the larger project,
the basis of the other methods used and characterises the
difference between the systems used in the smaller and
larger projects. Although it is primarily applicable and
essential to the larger projects, it is also useful in the
management, planning and control of the smaller project.
The use of the WBS is combined with a n approach to
planning which creates a hierarchy of plans, with differing
levels of detail, and a performance analysis approach to the
control of projects which meets the requirements outlined
previously for a good project control system. Ancillary to
these three key elements of advanced project management,
i.e. the work breakdown structure, a hierarchy of plans and
performance analysis, are the treatment of the larger project
as a portfolio of smaller projects and a formal change
control system to manage change. Therefore the elements of
-
204
advanced project management can be defined as follows:
1 Work breakdown structure - the use of a formal

structure to break the project down logically and
systematically into its component parts to:
(a) enable the planning to be done effectively by
defining the work to be done to complete the
project and to subdivide it into manageable tasks
that can be planned, budgeted and controlled;
(bj assign responsibility for the completion of these
tasks to internal functional groups or organisations and/or contractors, and thus to integrate
the organisation structure with the work to be
done, i.e. the work breakdown structure;
(cj design and integrate the control and information systems, with the work to be done andwho is
responsible for it.
2 Hierarchical planning - the use of a hierarchy of
plans, combined with a rolling wave approach to
planning the larger project, which can be extended to
multiple projects, the portfolio of smaller projects
and the very, very large or macro project.
3 Performance analysis - a performance analysis
approach to the analysis of data to produce simple to
understand, but more meaningful information for
management and control, which meets the criteria
outlined for a good control and information system.
4 Management of the larger project as a portfolio of
smaller projects - the treatment of the larger project as
a portfolio of smaller projects to simplify the management, planning and control.
5 Change control systems - an emphasis on the
systematic control of change actually to accept it as a
normal part of managing the larger project, and to
manage it.
Work breakdown structure

There are three interrelated stages to the application of the
work breakdown structure concept to the larger project:
1 The initial breakdown of the project into its component parts to define the work to be done and form the
basis for planning this work
2 The integration of the work breakdown structure and
the project organisation structure, sometimes now
termed the organisation breakdown structure.
3 The integration ofthe work breakdown structure, the
organisation breakdown structure and the control
and information system, together with the allocation
of funds to the work tasks defined.
The initial breakdown of the

breakdown structure
project into its work
The term work breakdown structure is simply a name for an

end-item oriented family tree subdivision of the projects,
products, or item to be built, components, work tasks and
services required to complete the project. It defines the
project, the work to be done, the organisation, and graphically displays the project It is very similar to the bill of
materials used in production planning.
The work breakdown structure is constructed by exploding the project into its component parts and services required,
with each of these components being further subdivided
into lower level elements. This explosion is continued until
the project is fully defined in terms of'what' is to be done to
complete the project, described as work breakdown structure
'elements'. Tasks of lesser complexity, or size, may require
substantially fewer levels of work breakdown. Although
primarily oriented towards discrete end products, software,
services and functional tasks may also be included, provided
each element is a meaningful product of a managementoriented subdivision of a higher level work element. This
work breakdown structure divides the work on the project
into manageable units of work, for which individual responsibility can then be defined. Figure 7.1 shows in arrow
diagram form, the activities involved in the design and
manufacture of a project. Figure 7.2 shows one way of
constructing a work breakdown structure for this contract.
In this example, the first level breaks down the overall
1 m[
U) lo
X U
>
TOY
project into the mechanical subassembly, the electrical and

instrument subassembly, basic design, final assembly,
general administration and overheads, and project management. In the next level the electrical assembly is broken
down into the electrical components, instrument components, basic design and the integration into the subassembly.
This then gives the lowest level of the work breakdown
structure of the project, i.e. the work breakdown structure
elements.
It is often possible to construct a WBS for a project in
several different ways, just as with a bill of materials for a
product, with a different number oflevels of breakdown and
numbers of elements in each level. It is thus useful for a firm
to specify a standard format for the characteristic type of
project it handles. One example of this is the 'Military
Standard, Work Breakdown Structures for Defence Material
Items' used by the US Department of Defense. This covers
the first three levels of breakdown for commonly procured
items and leaves the contractor to define the further breakdown of the project. Figure 7.3 shows parts of the work
breakdown structure for a ship system based on this
standard.
WBS and the project organisation structure

Thus the work breakdown structure is a formal method for
identifying the work to be carried out in a project. However
the project's global organisation structure is also a formal
structure which shows how the people and companies
involved are going to carry out the work Integration of the
work breakdown structure and organisation structure is
necessary in order to assign responsibility for the tasks to be
performed. This interrelationship can occur at any level of
work breakdown, but it is critical that this integration exists
at the level where work is actually carried out.
A project is thus broken down vertically, using the work
breakdown structure, into the lowest level work breakdown
elements and then the functional work involved in these
elements is integrated horizontally into the project organisational breakdown structure. The functional work represented
ELEMENTS OF ADVANCED PROJECT MANAGEMENT
21 1
by this intersection, as shown in Figure 7.4, is then a

complete project task which is the prime responsibility of an
organisational unit or manager. The name now given to this
task is a cost account.
A cost account is then a logical management centre on
which to integrate the organisation, the work, the manpower, the budget and the control and information system. It
is a self-contained entity, essentially a small project, with a
simple management structure, i.e. one functional manager is
responsible for it, and a cost account can be treated as an
informal or formal subcontract between the project manager
and this functional organisation or manager.
A cost account then has:
1 A single person responsible for it.

2 A formal specification of the work involved in it.
3 Its own workplan with key events or milestones
integrated with the overall project plan.
4 A manpower plan.
5 A time-phased budget.
Just as these factors are defined for the overall project, so
they are defined for the individual cost accounts. For
planning purposes these cost accounts can be treated as
small projects and are thus in turn broken down into the
normal activities as described in the chapters on planning
for the smaller project. Individually or in small groups, these
activities are termed work packages and when not completely defined as planningpackages, as shown in Figure7.5.
Each package then has the normal scheduled start and
finish, manpower and expenditure budgets. Work packages
are described in greater detail in the section on performance
analysis.
Attempting to portray the full complexity of this organisational form, which is but the fuller expansion of the matrix
organisation, on a conventional organisational chart is
difficult, if not impossible. Thus any form of matrix organisation, and in particular this kind of extension, is normally
shown in a matrix of responsibilities chart.
A responsibility matrix is simply a graphical form for
indicating who is to do, or be responsible for, the various
u a
.*
>
u.
s
81 2':
e m
e a mr =r
0
- X E P
urn.*,
0 -
214
component jobs making up a project. Figure 7.6 shows a

large-scale quadruled form on which jobs are shown on the
ordinate, and people, and or companies, on the abscissa.
Symbols can be drawn in the co-ordinates to represent the
type of work or responsibility of each person on each job.
This matrix can show who has responsibility for which
phase of the project what that responsibility is, and all the
jobs making up the project. It is easy to make up, permits the
individual to check what his responsibilities are, so that
nothing is overlooked and job overlapping is avoided. It also
shows whom he should consult on other phases of the
project and permits management to see at a glance who is
working on the project. It is essentially the complex organisation chart for this matrix form of project organisation.
Thus the matrix of responsibilities chart shown in Figure
7.6 can then be constructed to show the integration at points
of intersection of the work breakdown structure and organisation structures. Each organisation then is clearly identified
with the work for which it is responsible. Further subdivision of the effort may then be accomplished by the
appropriate organisation managers assigning work to subsidiary units, with subcontractors and equipment suppliers
also included in this matrix.
WBS and the control and information system

The WBS is thus used to define the work required to
complete a project and allocate responsibility for this work
to individuals and organisations. However, in addition to
this the WBS performs another function, no less critical to
the success of modern project management, which is to
establish the structure or framework of an integrated project
management information system. It can be used to integrate
the work to be performed, the organisation structure,
individual responsibility for the work, together with the subsystems for planning, estimating. budgeting, information
analysis, control and reporting into one logical integrated
project management system. This type of integrated system
is essential to the management of large-scale or complex
Arrow diagram
8.7 "eeks
C9.0~0
1-5 weeks
C-8 ueeks
r1o.m
f7.000
r\ E-10 weeks
'LJ 1 1 2 , o w
1-f
F-2 w e ~ k *
.LJ r 3 . o ~
weeks mJ-4
weeks
~ 8 . 0 ~ 0 ~ t h . o o o
Time Analysis
LP
TF
5
5
5
12
12
5 1 2 1 3 2 0
5
b 1 O l l
12
12
22
22
13
20
15
22
10
11
17
18
I7
I8
21
22
22
22
28
28
28
28
32
32
Activity
ES
LS
1
2
3
A
I
1
2
2
2
3
2
3
4
5
7
b
7
8
9
5
1
8
5
LO
2
5
b
1
8
9
10
C
D
E
F
C
H
5
h
7
8
7
I
6
4
EP
5
U
7
7
I
216
I
0
Figure 7.8 Project plan for simple example
Cri.Perh
1
0
Bar C h a r t
C r i . Path
Cri. P a t h
Cri.Palh
Cri.Parh
ELEMENTS OF ADVANCED PROJECT MANAGEMENT
217
projects and without it this type of projectwill inevitably cost

more or take longer to complete.
The intersection of the work breakdown structure and
organisation structure at the lowest level of the WBS is a
convenient and logical point to integrate all these systems.
The assignment of lower-level WBS elements to responsible
lower-level functional managers, provides a key automatic
integration point for all these subsystems and for the
management, planning and control of the project. The
expansion of these 'cost accounts' into work packages
extends this to more detailed work and these work packages
form the basic elementary building blocks of the project
management information system. The WBS then provides a
framework for consolidation of these systems for planning,
analysis and control for both project performance and
organisational performances.
The lowest level of cost accounts is located immediately
above the work package level and is normally made up of
several work packages. Figure 7.7 shows the work breakdown structure, cost accounts and work packages for the
example shown in Figure 7.8. Costs are accumulated,
progress measured and performance analysis carried out at
this 'basic' level. The measurement of Drogress is based on
completed work packages and work packages being worked
on in the veriod in auestion. Subiectivihi is limited to
estimating progress onAworkpackagks staried, but not yet
finished, and this considerably reduces errors in estimating
progress and quickly corrects any optimistic estimates.
These lower-level cost accounts should not be exorbitantly long, or there is a danger that there would be the use
of funds for later work for earlier work in the cost account
and the information system might have difficulty in differentiating this. At this basic level nine to twelve months is a
reasonable maximum length.
In addition to these lower-level cost accounts, there
should also be a hierarchy of higher-level cost accounts, as
shown in Figure 7.7, built up from these basic cost accounts,
each with the above information assigned to it, the higherlevel cost accounts being work breakdown structure elements.
It should thus be possible to cany out a performance
Activity
Estimate
Monthly Budget (1
6
4
5
3
1 2
7000
9000
3857
5143
10000
3750
5000
6000
3600
2400
12000
4800
3000
3000
9000
5000
8000
6000
75000
Figure 7.9 Project budget
1400
5600
2571
1250
5143
4800
2490
1286
3750
1250
2667
5333
6000
5600 12607 15114 14192
9836
6317
5333
6000
analysis on a summarised basis for segments of the project.

the project as a whole and for the organisations working on
the project.
The structure of this matrix hierarchy of cost accounts,
and the size and level of the basic cost accounts are
important decisions to be made at the outset of a project in
designing the project management information system.
These factors should be carefully considered to ensure that
the work will be properly defined into manageable units and
that functional responsibilities are clearly and reasonably
established. This is where the work breakdown structure is
an invaluable look at structuring the hierarchy of cost
accounts.
For example, consider the simple example shown in
Figure 7.7, which shows the basic level cost accounts, made
up of one or more packages, the higher-level cost accounts
andwork breakdown structure elements consolidating these
for segments of the project, and two organisational cost
accounts. The budget for each of these cost accounts can be
constructed as shown in Figure7.9 and performance analysis
carried out for each of them.
These cost accounts and the concept of representing them
as individual contracts are extended up the work breakdown structure and to the organisation structure to form a
matrix hierarchy of cost accounts or pseudo or real subcontracts. These basic cost accounts can be consolidated up
the work breakdown structure for analysis and reporting on
higher-level work breakdown structure items. This forms a
natural hierarchy of the work breakdown structure cost
accounts, each with its planned start and finish, time-phased
budget and analysis and reporting systems. They form a
structure of the planning and control system for the overall
project, with all subsystems integrated at each level in the
hierarchy. Each work breakdown structure cost account can
be considered as a separate contract and controlled as such,
with the ability to trace variation, changes and trends right
down to the lowest level cost account.
The basic level cost accounts can also be consolidated
horizontally, and all the cost accounts for each organisation
can be considered as a functional cost account or pseudo-
220
ADVANCED PROJECT M A N A G E M E N T
contract for that organisation. Again the organisation's total

work on the project will have a plan, budget, information
feedback, and be subject to analysis and performance
reporting.
The highest level cost accounts are the project itself on one
axis, and the performing company or companies on the
other axis. A full performance analysis can be carried out on
schedule and cost performance for all levels of cost account.
Any variance in the project can be traced down the work
breakdown structure to the basic level of cost account, the
work package involved and to the organisation responsible
for i t Performance of the companies, sub-organisations or
departments involved can also be analysed in a similar
manner. Reporting can be kept to the minimum and yet be
expanded to identify a cause and who is responsible for any
variance.
Managing the larger project as a collection of smaller projects
In order to deal with the problem of dimensionality in
managing the larger project, it can be treated not as a
collection of hundreds, thousands, or tens of thousands of
individual activities but as a collection of a much smaller
number of individual small projects, either work breakdown
structure elements or cost accounts. On the smaller project
there will only implicitly be one cost account, that is the
project itself. As the projects get larger and more complex,
the number of cost accounts used will increase, but so will
the value of this concept of breaking down a large project
into a number of smaller projects.
Not only does it make the larger project more manageable for planning and control but it also makes it easier for
the project manager to grasp it in his mind. A major
advantage is that the concept of treating these cost accounts
as small self-contained projects in their own right with
binding contractual obligations in terms of specifications,
schedule and cost can contribute to the commitment and
motivation of the people responsible for them.
Thus each work breakdown element or cost account can
ELEMENTS O F ADVANCED PROJECT MANAGEMENT
22 1
be treated as a separate small project but can be integrated in

the total project. The basic planning and control approach
and techniques described previously can then be used
effectively at cost account level. The senior management can
view the larger project in terms of a number of cost accounts
and not as a multitude of activities, thus making it easier to
visualise, manage, plan and control. This concept arising
out of the work breakdown structure methodology can be
combined with the hierarchical concept of planning to
enable the largest projects to be effectively handled by
human beings and not supermen.
Hierarchy of plans and the rolling wave concept

The problems described previously in planning the work on
larger projects are handled by using a hierarchy of plans, the
rolling wave concept of planning and the work breakdown
structure both to structure and to integrate the plans and
other systems.
A large project may require up to tens of thousands of
activities to plan and control the work adequately, and yet
most managers can only take in at a pinch something like a
maximum of 500 activities on any one plan. However, a plan
of 30 to 200 activities is easie~to follow and to visualise, and
most managers feel more comfortable with this size of plan.
Therefore although the global plan of the project may have
these tens of thousands of activities held that way in a
computer memory, it is so structured on a hierarchical and
modular basis that limits the size of plans that must be
reviewed and used by humans to these limits, i.e. 30 to 200
normally, and sometimes extending up to 500 activities.This
is not to say that larger plans are not produced, but more that
the tendency is to move to this basis as good practice.
Thus although on the smaller project a single level of
planning is quite sufficient, on larger projects a hierarchy of
plans with varying levels of detail is used as shown in
Figures 7.10 and 7.11. These are used formally and systematically to expand the work necessary to complete the project
into its component parts, using the work breakdown
222
structure, down to cost account level, and then to use work

packages to break down the individual cost account
Each single activity or small group of activities, in the
higher levels of planning is expanded into several activities
or complete unit plans at lower levels. The different levels of
plans are linked together and the events or milestones are
shown in Figures 7.10 and 7.1 1. In medium-sized projects
only a summary level plan and individual cost account
plans are required, while on larger projects three or more
levels of plan may become necessary.
Level 1 plan
A level 1 plan is essentially a summary plan which outlines

the project in skeleton form. It is used throughout the project
as a top management reporting and review document, but it
is also the only plan possible in the evolution or project
strategy phase, and is thus also the initial plan of the project.
At the evolutiona~ystages of the project it is normally only
possible to build this highly aggregated plan, with the
activities being large sizeable increments of work, such as
work breakdown structure elements or cost accounts.
Such a plan shows only the principal activities and the
maior milestones associated with the ~roiect.It is a broadscale plan embracing design, procurement, manufacturing
and. or construction. and commissionine. It contains the
first appropriate estimates o l the overac timing of each
stage, rough estimates of resources required, and permits
certain key points in the schedule and major interrelationships to be estimated approximately. It should show up
which critical parts of the total project should be emphasised and gives the first indications of requirements for
materials, and equipment delivery dates. In the initial stages
it acts as a strategic planning tool to establish the project
objectives and strategies and can be quickly constructed and
modified to show different ways of carrying out the project,
and yet allows key points to be located and major relationships determined However it is very coarsely scaled and on
large projects cannot be used for the complete integration of
the various project stages, which leads to the development of
/
/
Figure 7.10 A hierarchy of network plans

223
Level
2 plans
1r y ~ ~
,
Level
2 plans
----
1_1
'\
--
.:I-
I1
,'
I
I
I -
\
I
--
I
I
,
,
-I
'
. '
Figure 7.11 A hierarchy of bar charts
224
.- -
a level 2 plan. In one North Sea oil project, the level 1 plan
contained only 70 activities for several hundred million
pounds of expenditure.
Intermediate level 2 plan

This plan shows the broad sweep of activities with both
major and minor milestones included. It collects together
under one descriotion several detailed activities whenever
possible. It permits a detailed examination of the project's
structure to be made. and allows relationshi~swhich exist
between the various parts of the project td be seen and
studied. It defines the limits, between which individual
activities can move, without affecting project completion.
There may be a separate level 2 plan for every activity in the
level 1 plan.
In a level 2 plan the activities in the summary plan are
expanded and planned in more detail. It is a middlemanagement decision and control tool identifying functional responsibilities for items on the work breakdown
structure. That is, activities are generally cost accounts or
work packages, and not of a size to permit day-to-day or even
week-to-week scheduling and control of work, except perhaps in critical activities. The example of network shown in
Figure 7.1 is a level 2 plan. This may be a 'skeletonised' plan
of the project with strings of activities consolidated into
single activities. However, this plan should be in sufficient
detail to allow a certain amount of manpower planning to be
done, and to show the principal sequencing and interrelationship constraints on work on the project.
Where contracts are put out to tender, the amount of
planning possible is based on the amount of design engineering completed, and these factors influence the type of
contract used. To enable a contractor to bid for a fixed price
contract, the project must be sufficiently defined, that at the
very least a level 2 plan can be constructed for the proposal
or pre-tender stages. Otherwise the contractor cannot be in a
position to make a fixed price bid, unless he is tendering a
'package' engineering design, procure and construct plant,
more or less off the shelf.
226
Where the contract form is cost plus, normally the design

is not at such an advanced stage that a level 2 plan can be
constructed, and a summary plan is all that is possible.
However, every effort must be made to have at least a level 2
detailed plan at this tendering stage, particularly for the
design work, both to enable reasonably accurate prices,
resource requirements and completion dates to be estimated,
but also so that when approval is given, or contract awarded,
the project can be off to a quick start, with the correct
priorities.
It is easier to gain time with less resources at the early
stages in the project life, than at the end of the manufacturing or construction stages, when literally hundreds, or
thousands, of men may be involved. Sufficient planning
must be done in the initial plan to ensure that the critical and
near critical paths are determined before approval of the
project is given. The same principles apply whether the
client company or a contractor will develop the level 1 and
level 2 plans so that when the contract is awarded, engineering design and purchasing can immediately be started on
the critical items.
Cost account level 3 plans
The work required to complete the individal cost account is

detailed in the lowest level of planning. Cost account plans
are used at the lower levels of management to help them
manage the day-to-day. week-to-week work on the project.
The important factor is that someone must plan the day-today. week-to-week work to ensure good management and
control. This must take into account resource limitations,
interrelationships, time targets and be a basis for budgeting,
organisation and control of progress and cost.
This detailed planning can be tackled in the following
ways:
1 All the cost account plans can tied together into one
integrated detailed project plan often involving tens
of thousands of activities.
2 Each cost account is represented by one activity in
the lowest level of completely integrated project plan
and the detailed cost account plans are separate

entities.
3 The lowest level of completely integrated plan includes for each cost account the principal events or
milestones and thus the cost account is represented
in this plan by several activities. The individual cost
account plan explodes this into more detail but
retains these principal events or milestones as its
structure.
Figures 7.12 and 7.13 show one way of carrying this out
Figure 7.12 shows the cost account plan for the electrical
drawing office of the example shown in Figure 7.1. It
combines all the planning information into a self-contained
plan for this one small project. Figure7.13 on theother hand
shows diagramatically the consolidation of this cost account
plan into higher-level plans up the work breakdown structure. The same principle is used to consolidate from the cost
account plan up the organisation breakdown structure. At
each level the plan shows who is responsible for the work,
the work plan, the manpower plan and the budget in the
detail appropriate to that level.
Different techniques may be used with different levels of
planning: a bar chart or milestone chart commonly being
used in a level 1 plan, CPM/PERT and bar charts in a level2
plan, and bar charts and/or milestone charts in level 3 plans.
However, one or all of these techniques may be used for all
levels of planning.
The rolling wave concept of planning

At the start of a project an initial level 1 summary planwould
be constructed outlining the total project. In many longer
projects it is often not practicable or possible, because of
lack of the necessary information, to plan at the complete
intermediate level or in particular to a level below cost
account level. It may only be possible to plan in detail the
early stages of the project because the information for the
planning of the later stages of the project will be generated
by this early work
Project : Nodular Robot

Company : A.N.
Dept.
Other
: Design
C.A. Name
C.A.
Manwr
: E l e c t r i o d D.O.
- Title,
IU manager
YBS Code: 1012104
OBS Code:
Name: F.L. Harrison
T o t a l Cost: C244.520
Total d i r e c t hours: 9600
- Inchstones
Figure 7.12 Cost account plan

228
231401
Pm?.=t
v*Yu Brnl C.l. la*
r.*.
Omcr
rimer
C.".
:
Tltlr:
a-:
~ l . ~ t r s ~Dr. L
i
Y B I W: tO3Z(Dd
0s so.,
7.L.
."-r
R.".
2 M.,
IY*OOllrt
D
HDyn
liwnrL
9.uo
& *in',
%la
xw
a7
5.m
,.XU
55.m
,i.XIO
3,s"
ku*u*
l l n i r LaBrui
toll
nar' mno. Ira"? at,t,rd.

anll*m/oMI
s.-,ee*
71,MO * . M a
2.m
2.010
1.m0
7a.ora
5.m
8"-o"rmc%
c.uia
~mtn..trs:,v.
arm
70-
lDirl
6%-o:
:-.n
PLm
3i.m
?d.MO
I5.m
7o.ao 7a.oaa
~a.m
DL,UO Bi.810
,r.BU)
- ,"a*,,--
Tntd ?"d:
ca+,%s
Figure 7.13 Hierarchy of plans/cost account version

229
r><to-
,8arrhS"
230
Thus it may only be possible to plan the total project in

sizeable increments ofwork, at the early stages of the project
and only the early stages in detail. Then as more information
becomes available the middle stage of the project can be
planned in detail and later on the final stages. This can be
likened to a rolling way moving from left to right, from the
start to finish of the project, with the activities behind the
wave planned in detail, say to work package level, and the
activities in front of the wave being planned at a coarser, say
cost account, level. That is, behind the wave activities consist
of work packages and in front of it of cost accounts. This
same principle can be extended to cost accounts with the
early stages planned as work packages and the later as less
well defined and larger planning packages.
Although this rolling wave approach is often necessary for
the larger project, the detailed planning of the later stages of
the project must be done with sufficient lead time to ensure
that the correct decisions and resource allocations are made
in an integrated manner.
Multi-project planning and control
Where a company or a project manager is handling several

projects at the same time, and particularly where resources
have to be shared between projects, it is essential that the
planning and control of these mutliple projects be integrated. The portfolio of projects being managed must be
then looked on as one project Thus many junior project
engineers, who are handling several minor projects, are in
fact handlingwhat is, for them, a largerproject, with many of
the problems outlined in this chapter. Common examples of
this situation are:
1 An owner company, for example, a major chemical
company, which has several major projects on the go
at any one time, each atvaricus stages of its life cycle
and each probably with a different contractor.
2 A contractor who has several projects each with a
different client company.
3 A factory with a mix of small and medium-sized

projects, using its own resources and occasionally,
when handling large projects, using subcontractors
or contractors.
4 A local authority with a large number of minor
projects and a few medium-sized projects, using a
mixture of direct labour and contractors.
5 A junior project managedengineer responsible for
several minor projects, primarily using in-company
resources, in a company where projects are secondary activities.
The major problem in such situations is the sharing of
resources between the following projects:
1 Key management personnel, i.e. project managers,

2
3
4
design managers and supervisors, construction

managers, and foreman.
Design engineers, particularly specialists.
Drawing office personnel.
Directly employed construction labour, in the last
two examples.
Floating construction labour availability on a
regional basis.
Where resources have to be shared, it is necessary to carry

out integrated planning of multiple projects in some form or
other. The total availability of these key resources become
constraints which must be recognised in the planning of the
individual project This integrated planning could be carried
out in two ways: the plans for all the projects are integrated
into one big multi-project plan, or the subplans for each
project for only the resources being shared are integrated in
one multi-project functional plan.
When the projects are small there are few problems in
integrating the plans for all the projects into one and in
carrying out manpower planning. Simply by using artificial
activities to impose delays on the starting of each project, or
each stage of a project, multi-project planningcan be carried
out. Another level is added to the top of the work breakdown
structure, for control purposes, which represents the multi-
232
project, with the first level of breakdown being the individual projects. Even with larger projects it is physically
possible, using modem computer systems, to integrate these
plans together. These systems have multi-project options
which can deal with the individual projects with its own data
file, but which can also consolidate these files and plans
together into one multi-project plan.
The principal difficulty with integrating large projects is
that the total plan becomes very large and cumbersome and
in practice it is not really necessary. The only integration
between projects that is required is where resources are
shared. Therefore it is simple, quicker and more flexible to
do this integrated planning only for those functions which
are shared. For example, a firm may be handling three
projects at the same time, each requiring an average of 40
draughtsmen for five months, that is, a total load of 600 man
months, with a peak requirement for each project of, say 70
draughtsmen. If it only has 90 draughtsmen available and if
it attempts to push all three projects through the drawing
office at the same time. there will be total chaos and infighting betmeen project managers for priority for their
vroiects.
If no other drauehtsmen can be obtained. for
"
example, by subcontractin; or hiring temporary labour,
there is no way that the drawing office work on these projects
can be completed in less than approximately six-and-a-half
months.
If it is left to the system to muddle through there will be
complete disorganisation and it will probably take considerably longer to complete the work The drawing office
plan for all three projects has to be constructed on a logical
and top management-imposed priority basis, with a major
constraint being the availability of draughtsmen.
Individual project plans would be constructed and the
drawing office work for all three projects integrated into one
functional plan, which would then be adjusted to level the
drawing office resource requirements to that available, by
establishing a priority for each project. The rest of the work
on each project could then be planned around the drawing
office constraint by inputting the necessary scheduled starts
and scheduled finishes for drawing office work determined
in the integrated drawing office plan. It may then be possible

to expedite the highest priority project through the work
prior to the drawing office, such that the bulk of its drawing
office workload arrives earlier than vreviouslv vlanned If
this can be done the planning a n d balancing process is
reveated so that the comvletion dates for the drawing office
work would be staggerei but not significantly delayed.
The same principle can be applied to any function which
has its resources shared between projects or parts of a
project.When a large number of resources is shared between
multiple projects, it will be necessary to carry out fully
integrated planning as there will be too many complex
interactions between projects and resources. If this was
attempted on an individual function basis it would require
many iterations and people would become confused. Automatic computer-based resource levelling would have to be
carried out with project priorities set by the planner, with
limited man-machine interaction. This would produce a
final integrated multi-project plan which makes sensible use
of priorities to achieve the objectives of all the projects and
the company as a whole.
Multiple projects in the minor project situation
The junior project managedengineer responsible for several

minor projects in a situation where projects are of secondary
importance is a somewhat special case of multi-project
planning and control. i n the other example quoted, projects.
if not the primary activity of the organisation, are of major
importance and thus organisation, systems and resources
are project-biased. In this example the minor projects are of
very much secondary importance. The contributing functional departments' primary activities are probably not
projects, e.g. they may be maintenance or manufacturing,
the organisation's information systems are not projectoriented and the project manager is probably in a weak
matrix organisation and is acting more as a project coordinator whose responsibilities far outweigh his authority.
In such situations the contributing organisations are
primarily concerned with their mainstream activities and
only allocate to projects resources not required for these
234
activities. These projects are then normally under-resourced

and only worked on intermittently, with resources being
withdrawn every time there is a surge in demand from the
mainstream activities.
These projects often take a long time to complete and are
carried out inefficiently, with parts of them perhaps being
completed quickly, if one department has spare resources
and other parts taking forever, if the reverse is the case. Work
is unco-ordinate4 both on the total project portfolio and for
the individual project, and life is full of frustrations for the
project manager.
The same principles that apply to the larger project and
the multi-project situation, as outlined above, apply equally
if not more so to this situation. The portfolio of minor
projects must be considered as one larger project, with
minor projects being completed and started as required, but
all integrated together as follows:
1 A work breakdown structure must be established,

with the first level of breakdown being the individual
minor project. Often no further breakdown will be
required.
2 The work breakdown structure and organisational
breakdown structure must be integrated and cost
accounts established, albeit in many cases these cost
accounts may be quite small.
3 A hierarchical work plan must be made for the total
project portfolio, with the first level plan made up of
individual projects.
4 The overall manpower plan and budget must be
established.
5 The individual contributing organisations' work
plan, manpower plan and budget must then be
established and agreed.
6 The control system must work in two dimensions, i.e.
up the work breakdown structure for the projects and
up the organisational breakdown structure for the
contributing organisations.
7 A matrix of responsibilities can be established.
Each contributing organisation will then know its total
resource commitment to project work, which can then be

compared to the forecast resources available in the light of
past experience. If project requirements exceed the expected
resources available then the work on the portfolio of projects
would have to be rescheduled and balanced between departments. Once a resource commitment was agreed with each
department and a project plan completed based on this confirmed, this levelwould be considered a contractual commitment. In some cases this may take the form of a minimum
contracted resource level, with a variation above this level
dependent on mainstream levels. Any planning would have
to be adaptive, but co-ordinated between projects and
departments to take advantage of increased resources available.
The project manager would then be able to manage his
portfolio of projects and co-ordinate the work on individual
projects and his total portfolio. He will be able to balance the
resources available with the resources required for any mix
of project work For example, if departmentA has resources
above the minimum agreed level, which projects should it
work on, what demands will it make on other contributing
departments and how can they be met? He will better be able
to allocate priorities, identify bottlenecks,forecast completion
dates and accept or reject new minor projects, or conversely
forecast completion dates for them.
The monitoring of individual contributing departments'
actual resource commitments and performance will enable
him to exert authority over them and to control their
activities. For example, a department may have agreed to a
minimum resource allocation to projects of6,000 man hours
per month, with an expected average of 8,000 man hours.
Control information may show it committing only 4,000
man hours per month, contraryto an agreed plan and senior
management can then be brought to exert pressure to
increase this, or conversely replan the work and forecast
completion dates based on a new contract. Similarly, if 6,000
hours have been worked and only the equivalent of 4,000
man hours of budgeted work have been completed. Thus
even the most junior project managerlengineer in this
situation is faced with the problem of handling the larger
236
ADVANCED PROJECT MA N A G E M E N T
project and can benefit from the use of this other dimension
of advanced project management.
Planning the very large project
The planning of the very large project, sometimes termed a
giant or mega-project, is more critical than ever to its
efficient management and control, but has considerably
greater problems than that of planning the medium to large
project. There are two extremes in the management, organisation, planning and control of such projects with many
variations between them. These are:
1 Complete centralisation, involving the integration of
the planning of all areas of the project.
2 Complete decentralisation, with parts of the project
being planned separately, and integration occurring
only at a macro level, or where resources are shared.
It is physically possible using modem computing systems to

use a centralised project plan, but if integrated planning is
attempted in any detail it requires anything from 3,000 to
75,000 activities. Such a vast plan would be complex,
cumbersome, costly, very difficult to create, constantly
requiring to be changed and almost impossible to keep upto-date. The information necessary to build the plan for the
later stages would only become available over a very long
period of time. The state of the art and science of planning
would undoubtedly vary considerably among all the companies involved, and attempting to impose one detailed
system would be very difficult and lead to a lot of resistance
from many of those involved. In addition, the normal
difficulties in achieving good communication and coordination are also compounded in these projects, when
they are centralised and made more difficult by the many
layers of management that are often necessary. Major
problems are the time necessary to get decisions made and
the sheer superhuman task of handling the mass of data
involved. Present-day practice is to move some way towards
decentralised management of mega-projects,and this section
of the book outlines only one of the ways that this can be
done.
A very large project can be viewed as an aggregation of a
number of separate medium to large projects, and is a
special case of multi-project management, planning and
control. In the following discussion, for the sake of simplicity,
avery large
be termed the 'programme' though in
- projectwill
. .
practice programme and project are interchangeable nimes.
Normally, the owner company, or a principal project
management company, would carry o u t the overall programme management, planning and control, involving the
integration of every contributing company's work, and the
bulk of the work on the programme will be carried out by a
number of construction and manufacturing companies who
will have contracts for the various projects making up the
programme. Each of these companies will have their own
project management staff and will be supervised by either or
both of the owners and programme managers' staff. These
companies will view their contracts as completely separate
projects, and the programme managers' staff will supervise
them, both as separate projects and as part of the interlinking programme.
It is in the planning of these very large projects that the
hierarchical and rolling wave principles of project planning
are essential. The number of levels of planning used can be
considerably greater than the three outlined previously but
not all would have to be integrated. The interrelationships
required between the individual project plans making up a
programme are normally limited to key major decision
points or milestones, and the programme plans need only
include these to give the necessary overall integration of the
programme. The number of levels of plans and their
integration will vary from one programme to another, but
one example of such a hierarchy is as follows:
Late1 I programmeplan This will be a skeleton plan of the

programme showing only the key programme decision
points or milestones, with most of the activities being
individual projects or contracts.
238
Level 2 programme plan This would be the consolidation of

the individual level one project plans into one integrated
programme plan, and would identify the major key programme interrelation points or milestones between projects.
LevelZ/3programmeplar2 It may be necessary in some cases,
when there is a large amount of interrelationship between
projects, to use another intermediate integrated plan,
showing the minor key programme interrelation points or
milestones between projects. This would be the consolidation
of individual project plans, somewhere between levels 1 and
1
Individual project plans

Below the above levels. separate project plans can be built
for each project using hierarchical principles as described
previously, and having the programme's major and minor
key milestones shown in them. The individual project level 1
and level 1/2 plans will be part of this hierarchy or project
plan, and also part of the above project plans.
Progress between milestones can be transferred from the
individual project plans to the higher level programme
plans to monitor and integrate the total programme. When
each project is handled by separate companies each with its
own manpower resources, this system can work effectively.
Integrating the manpower plans of the individual projects
into one programme manpower plan is only required where
there are resources shared between projects, for example, in
the owner or programme management company, the regional
availability of labour, say, in a construction camp, or where
space and accommodation are limited as on an oil platform,
or where the project is nearing completion. In such cases
integrated functional planning must be carried out for the
programme as a whole, as discussed in multi-project
planning.
The rolling wave concept of the expansion of the programme plans into greater detail and finally into separate
detailed project plans is also important in the planning of
large projects; earlier projects producing information to
permit the expansion of later project plans into greater

detail.
The separate project plan could be constructed using any
acceptable technique with which the individual company is
familiar, as long as it met the overall effectiveness criteria
previously outlined. Where any contributing company's
system does not meet these basic criteria, it is often necessary
for the programme management staff to impose some level
of planning technology on a contributing company. The
level of planning technology imposed should be the minimum acceptable level as attempting to move a company's
personnel from working by the seat of their pants to interactive computer-based detailed planning in one step is
normally impossible. Simple bar or milestone charts are
quite acceptable, normally easily implemented, and can
identify the key milestones in the programme plan in these
individual project plans and permit their integration into
the overall programme. Uniformity of planning would only
be necessary at the levels of the integrated programme plans,
which would be constructed by the programme staff.
If the work breakdown structure concepts are used to
explode the programme into its component projects and
also to explode the individual projects into their component
elements, control and reporting can be implemented as
described in the following chapters. This would permit the
consolidation of performance analysis and the tracing of
deviations from plan and budget down to the individual
project and to the individual detailed activity.
This method of management planning and control would
force much of the necessary decision-making, communication and co-ordinating down the line to the level where work
is being carried out and lead to a simple management
structure with fewer levels of management The organisation
structure, or structures, involved in large projects is normally
a mixture of functional matrix and divisional organisations.
In practice the forms of organisation structure used will be
dynamic and change over the life of the project For
example, a very large project may start with a functional
organisation, move to a divisional organisation, then to a
matrix a n 4 finally, may contain all three types, both in the
240
global organisation and in the component parts of it.

A very important factor in the management of programmes, involving hundreds or even thousands of millions
of pounds of expenditure, is the forecasting and management of the cash flow. The forecast cash flow of all the
individual projects has to be consolidated into the overall
programme cash flows. Then as progress and performance
vary, the cash flows have to be planned and controlled, that
is, managed. With the present-day rate of inflation, this can
involve many detailed calculations as described in the next
chapter, and computer programmes or financial models
have to be used to handle this. These forecasts and models
have also to be integrated with the various levels of programme plan.
8 Performance Analysis
Modern methods of organising project control systems,

sometimes termed performance analysis are based on the
use of:
1 Three different types of control centres to control
progress and cost, with differing modes of analysis.
2 The work breakdown structure to provide a framework to integrate the subsystems and to consolidate
reporting.
3 The earned value concept for work done.
The most sophisticated application of these concepts is seen
in the US De~artmentof Defence. cost schedule control
system criteria (C/SCSC) and cost performance reporting
svstems (CPR). These are ~rimarilvfor the US defence
Gojects, but the concepts aLd tech;iques used are just as
applicable to manufacturing and construction projects in
industry.
Control centres
There are three types of control centre required for effective

project control, each involving different methods of analysis:
1 Work packages: essentially discrete tasks, or activities
as on a bar chart or arrow diagram, on which
analysis is by relatively straightforwardbudgetversus
actual variance measurements, as in PERT/COST.
2 Overheads: on which analysis is based on an expenditure rate over time.

3 Cost accounts: the aggregation ofworkpackages into
separate sections of the project, the largest ofwhich is
the project itself, on which a full performance
analysis, forecasting of final cost and completion
dates are required.
Work packages
The basic building blockofthe modern planning, budgeting
and control system is termed a 'work package'. A work
package is simply a discrete task, activity or job, which has
an observable start and finish, and an end product of some
form. It is normally related directly to one level of the project
plans, be they bar charts or networks. In the simple example
shown in Figures 7.8 and 7.9, each activity could be considered as a work package. A work package can be a single
activity, a cost hammock covering several activities in series,
or several activities progressing in parallel. For example, a
work package can be the excavation, setting of shuttering
and reinforcing, pouring of concrete and placing of a single
vessel, or piece of equipment. Alternatively,it could be those
activities for all the equipment on one area of a project, or it
could simply be the excavation, or any other individual
activity in the sequence, for all the equipment on the project,
or in one area of the project
Each work package is thus a significant amount of work,
which is large enough to simplify the estimating, budgeting
or collection of costs and other information, but which is
also small enough to pick up deviations before they become
dangerous. This concept attempts to achieve the objectives
and the compromise outlined at the end of the last chapter.
Analysis of progress on work packages is by simple
variance analysis, as carried out in PERT/COST, and this is
quite adequate for these short timespan activities. However,
work packages are different in concept from the activities
used in PERTICOST, and are only the base element of the
project control system. They are not simply cost centres, but
are more information or control centres. They are in effect
PERFORMANCE ANALYSIS
243
small pseudo-subcontracts between the project manager

and the individuals contributing to the project They describe the work to be done, are the assigned responsibility of
one individual, have an estimated cost and a time schedule,
and their completion gives an objective means of monitoring progress and cost.
The size of the work package used is not tied to the size of
the activity necessary for physical planning purposes, but
can vary between functions. The key factor of the work
package concept, from the point of view of monitoring
progress and cost is that they should be relatively short.
Work package durations should be measured in terms of
weeks rather than days or months. With short-term work
packages, overall progress and cost analysis can be based
mainly on completed work packages. Subjective estimates of
progress are limited to 'open' work packages, that is, work
packages that are started but not yet completed. Ideally, as
control analysis is normally based on monthly reporting,
most work packages should be from one to two months'
duration. No attempt should be made to achieve this artificially and workpackages should be natural subdivisions of
the work. When work packages are longer they should be
subdivided by objective milestones, or 'inchstones' with
separate estimates and schedules for each segment. On these
short work packages, there is little need to carry out an
analysis of cost and progress until they are completed. As
they normally span two, or at most three reporting periods,
an optimistic subjective assessment of progress in the first
reporting period is quickly corrected in the second. This
short timespan of a work package is essential to reduce the
amount of subjectivity and error in performance reporting.
Thus project performance analysis at work package level
is based on variance analysis of completed work packages,
and analysis of performance on larger segments of the
project and for the total project can be based on this
foundation. However, one cannot wait until a project or
substantial segments of it are completed before carrying out
an analysis of performance. A project can be made up of
anything from 30 to many thousands of work packages and
variance analysis of completed work packages is inadequate
244
to give a clear picture ofwhat is the overall project performance, or to forecast the final cost andcompletion date. This is
where simple PERTICOST fall down and where the more
sophisticated 'performance analysis' based on 'cost accounts' and work packages, can give more effective control
of projects.
Overheads
Direct costs, such as labour, materials and subcontracts can
normally be easily allocated to work packages. However,
many other activities can exist on a project which are more
general or s u ~ ~ o r t i in
v enature. and cannot be identified as
gelonging to ipecific work packages. They cannot be easily
and logicallv allocated. and are termed indirect or overhead
costs. The ailocation df these overhead costs is a problem
area in project management, just as it is in production
management, and if wrongly handled it can distort the
management decision
and make control of a project
difficult. It is here that accounting for management control
often differs from standard acco;nting
The traditional accounting practice is to endeavour to
allocate or apportion these costs to work packages based on
some criteria of the presumed benefit the work packages
receive from these indirect activities. Unfortunately, there is
often no direct link between work packages and these
overhead costs, and the criteria used can be artificial or
arbitrary. An overhead rate for the overall organisation is
sometimes derived on an annual basis of the proportion of
total manpower used, or as a percentage of total expenditure.
To do this, estimates are made of total annual company
activity and the total overhead cost, and a simple division
used to establish an application rate for overheads.
This can result in an arbitrary allocation, and may be
necessary and satisfactory for historical or proprietorship
cost accounting, but is generally useless and counterproductive for management control purposes. Many of the
sources of these costs are not activities which good or bad
performance on work packages can influence, or vice versa,
nor can the project management influence them, as they
245
have little responsibility or authority over them. To achieve

effective management control in project work, overheads
must be treated in a different manner than in traditional cost
accounting.
Overheads in project work can be divided into two
classifications, and each must be handled differently,that is:
1 More or less direct overheads, which can be allocated

or apportioned in a logical manner.
2 Indirect overheads.
Some overheads are more or less direct overheads and can
be directly traced as supportive effort to particular work
packages, or cost accounts, and are dependent on, orrelated
in direct proportion to the performance on them. These can
be logically allocated on a meaningful basis to particular
work packages, or may even form separate work packages
within cost accounts, each with its own budget which may be
based on a percentage of the related work package cost. For
example, a design department may be working on three
projects, and the departmental overheads can be split
among these projects and have some meaning.
Other overheads cannot be directly traced to individual
work packages, or cost accounts, for example, central
company overheads, project management or general
administrative costs. It is wrong, from a management point
of view, to allocate these arbitrarily to work packages or cost
accounts. These costs must be separated from that which is
measurable, to avoid distorting performance analysis.
Separate budgets can be made for these indirect overheads, and allocated to the project as a whole, or to discrete
segments of it. For example, the budget for the project
management costs can be allocated to the overall project
and the budget for construction management allocated to
the total construction phase. Generally these overheads are
time based, and are simply budgeted as a cost per period,
though in fact the level of expenditure may vary from period
to period. Measurement is on the basis of an expenditure
rate over the passage of time, and if the project is extended,
these costs are unavoidably increased.
246
Cost accounts
Carrying out a control analysis of a relatively long ongoing

activity, such as the project itself, or a major part of if is
entirely different from the simple budget versus actual
variance analysis for short span activities, that is, work
packages. One difference is that for effective control analysis
there must be a reliable measurement of progress that is
percentage complete, or to give it another name, the 'earned
value' of work done.
In practice this tends to be a highly subjective estimate,
made with'rose coloured spectacles', unless some systematic
method is used. Many projects, and segments of projects,
have had optimistic estimates of progress made until these
estimates are obviously shown to be widely inaccurate, and
by then it is too late to do anything about it. These projects
tend to be reported 'on schedule and budget' up to 80-90 per
cent complete, at which stage they stay that way until the
actual work catches up with these estimates, the project
completion date is extended, and the budget over-run.
Modern methods of analysis endeavour to limit the amount
of subjectivity, that is, opinion, used in the measurement of
progress and thus make control action possible at an earlier
stage.
The type of information the project manager needs to
have for effective progress analysis and control are objective
answers to the following questions, which can be called
'performance analysis'.
1 Are we on schedule, ahead of schedule, behind

schedule, on the project as awhole and the individual
parts of it? If we have a variation from schedule,
where did it occur, why did it occur and who is
responsible for it? What effect will it have on the
other parts of the project, and what can we do about
it?
-.
2 Is the work being completed to the budget estimate,
or less than the budget estimate, or is the work costing more for the project as a whole and the individual
parts of it? If we have a variation from budget, where
did it occur, why did it occur, orwhat caused it, who is
247
responsible for it, and what can we do about it?

3 Are we going to be on schedule, on budget, or over
schedule and over budget? That is, what is the trend
of progress and cost? It is vital to become aware of
trends at an early stage or there is little possibility of
influencing them. When work progress or cost is
obviously varying, it is normally too late to do anything about it The secrets of real control are to pick
up trends as soon as they start to occur and to be able
to do something about them.
4 Thus project management needs to be able to forecast the final cost, and the completion date for the
project and for the individual parts of i t
5 We are also concerned with monitoring against plan
and budget the rate of build-up of cost and progress.
This is analogous to the acceleration of the project,
just as cost versus time is essentially the velocity of
the project. (It is in this that S charts are an effective
tool.)
On small projects, the project manager will only want this
full 'performance analysis' for the project as a whole, with
the simpler budget versus actual variance analysis for work
packages. However, as projects get larger. this limited
analysis becomes too aggregated and does not give sufficient
control over the project. The project must then be broken
down into segments, forwhich a full performance analysis is
also carried out in order to answer the above questions for
each segment. Performance analysis for only the complete
project is too insensitive and slow to react to wayward
variances on any part of i t These segments can be called
'cost accounts'. In a small project there may be only one cost
account, the project itself, on slightly larger projects there
may be only a handful of cost accounts. In very large
projects, hundreds of cost accounts may be required to
provide effective control.
These cost accounts are key ~ o i n tfor
s the management's
planning analysis and control workon a project.xll parts
of the proiect
- - information system must be integrated at these
points, including budgets: schedules, work-assignments,
of
248
organisational responsibilities, cost collection, progress

assessment, -performance analysis, problem identification
and correctiGe actions.
Methods of analysis
The methods of analysis that should be used at cost account
and higher levels of the work breakdown structure and the
project as a whole, are based upon earned value concepts
and S chart analysis. The earned value concept is based on
the measurement of the budgeted value of work actually
carried out, and its comparison with the budgeted value of
the work that should have been carried out and what it
actually cost. The concept has been crystallised and defined
in the cost/schedule control systems criteria. Previously a
project manager could measure progress against a predetermined schedule and measure actual cost for work
performed against the budget. The problem was that the
project manager could not determine, for the money spent,
whether or not he was getting the progress he should have
obtained; how much should he have spent for the progress
achieved, and how did this actually compare with what he
actually spent? Often he tried to evaluate value by making a
subjective estimate, usually subconsciously based on the
percentage of funds expended. The success of the modem
methods of project control is based upon removing this
subjectivity.
This method of analysis uses three basic terms with which
the project manager should become familiar. These are
partially new concepts, and partially the formalisation and
'jargonisation' of previous methods. The terms used are:
1 BCWS - the budgeted cost for work scheduled.

2 BCWP - the budgeted cost for work performed.
3 ACWP - the actual cost of work performed.
Using these three data elements, it is possible to evaluate
quickly for each lower-level cost account and for the hierarchy of cost accounts:
249
Its cost variance

= BCWP-ACWP
Schedule variance in cost terms = BCWP-BCWS
Cost performance index (CPI) = BCWPIACWP
Schedule performance index in
monetary terms (SPI)
= BCWPIBCWS
5 Estimate to complete the project
(ETC) (where BAC = Budget = (BAC-BCWP)/
at completion)
CPI
6 Estimate at completion, that is,
the forecast cost (EAC)
= ACWP+ETC
1
2
3
4
Budgeted cost for work scheduled (BCWS)

BCWS is simply another name for the time-phased budget
against which performance is measured for the project and
for the individual cost account. It is normally determined for
each individual cost period and on a cumulative basis,
though in small projects it may only be determined on
cumulative basis. For any given time period, budgeted cost
forwork scheduled is determined at the cost account level by
totalling the budgets for all work packages scheduled to be
completed, plus the budget for the portion of in-process
work (open work packages) scheduled to be accomplished,
plus the budgets for the overheads for that period. Figure 8.1
shows the BCWS for each period for the sin~pleexample.
Budgeted cost for work performed (BCWP)
The budgeted cost for work performed consists of the
budgeted cost of all the work actually accomplished during
any given time period. This can be determined by individual
time period, and on a cumulative basis. At the cost account
level BCWP is determined by totalling the budgets for work
packages actually completed, plus the budget applicable to
the completed in-process work with open work packages,
plus the overhead budget.
The major difficulty encountered in the determination of
BCWP is the evaluation of work in progress (work packages
which have been started but have not been completed at the
time of cut-off for the report). As discussed previously, the
nonrhly Budget (L) i n Period
Control Cenrre
1
c o s t Accovnt
No. 1
work parkage A
Total
7.000 5.600
- 1,400 - - - - 7,OM 5.600 1,400
-------
Cost A C C D YN~O~
. 2
Work package B
Work package E
Taral
9.W
3,857
5,143
12.000
4.800 4.800
2,400
- ---.
21 ,WO
3,857 5.143 4.800 4.800 2,4W
- - --- -- -
c o s t ~ c c o u n tNO. 3
10,000
3,WO
3.750
13.000
3.150
Work package D
6.000
3,600
Vork package G
Work package H
5.000
Work package C
Work package F
Total
5.000
1.250
3,000
-- - - - - - 5 . W
-- - ---4,250
cost Account NO. 4
Total
9,020
2,400
2.571
5,143
1,286
- - - - - 3,750 1.250
-20,WO
3,600 4,971 5,143 5.036 1,250
-- - - - - - -
C o s t Account No. 5
Work package I
Vork p a c b g e J
Total
8,WO
6,000
2,667
5.333
-- - - - - - -- - - - - - 14.000
2.667
5,333
Department A
C o o t Account No. I 7,WO
C a s t Account No. 2 21,000
5,600
1,400
3,857
5,143
4,800
4.800
2,400
3.750
3.600
5.000
4,471
4,250
5.143
5.036
1,250
2.667
Department B
Coat ~ c c o u n f110. 3 1 3 , m o
C o s t Account No. 4 20.W0
C a s t Account No. 5 14,000
--
Total
Total Project
5,333
-- -- -47,WO
7,350 9,971 9,393 5,036 3,917 5.333
-- - - - - - 75.000 5,600 12.607 15.114 14,192 9,836 6,317 5,313
--- --- --
Figure 8.1 Manual calculation of project budget for cost

accounts
25 1
use of short span work packages, or the establishment of

value milestones within work packages will significantly
reduce the work in progress evaluation problems, and the
procedures used will v a q depending on work package
length. For example some contractors prefer to take 50 per
cent of the BCWP credit for a work package budget when it
starts and the remaining 50 per cent at its completion. Other
contractors use formulae which approximate the time
phasing of the effort, while others prefer to make physical
assessments of work completed to determine the applicable
budget earned. For longer work packages many contractors
use discrete milestones, with pre-established budget, or
progress values to measure work performance. The method
used largely depends on work package content, size and
duration. However. the use of arbitrary formulae should be
limited to very short work packages. Figure 8.2 shows the
manual calculation of BCWP for period 3 in the simple
example.
Actual cost of work pegormed (ACWP)
This is simply the costs actually incurred and recorded in

accomplishing the work performed within a particular time
period. These may be collected at cost account level, though
more commonly they are collected at work package level, or
individual activity level, depending on the size of the
activity.
Cost/per&ormance analysis
Using these three elements of data, it is possible to carry out

a n analysis of performance which integrates schedule and
cost Combining this with the work breakdown structure
permits this analysis to be carried out at will, for any part of
the project, for any contributing organisation and for the
project as a whole. O n larger projects, these comparisons are
carried out for the latest time period, and on performance to
date, that is, on a cumulative basis, whereas on smaller
projects they are sometimes only carried out on a cumulative
basis.
Estimate
X
Complete
Cumulative
BCWP
Previous
BCWP
Cost Account No. 1

Work package A
Total
Cost Account No. 2
Work package B
Work package E
Total
Cost Accounc No. 3
Work package C
Work package F
Total
Cost Account No. 4
Work package D
Work package G
Work package H
Total
Cost Account NO. 5
Work package I
Work package 3
Total
Department A
Cost Account No. 1
Cost Account No. 2
Total
Department B
Cost Account No. 3
Cost Account No. 4
Cost Account No. 5
Total
Total P r o j e c t
Figure 8.2 Calculation of budgeted cost of work

performance (BCWP) for period 3
BCP f o r
Period
253
Cost variance shows whether the work done costs more or

less than was estimated, and schedule variance will give a
cost measure of how far behind, or ahead of schedule the
cost account, work breakdown structure item or the project
is. Estimates to complete (ETC) and estimate at completion
(EAC) produced by this method are mechanistic methods of
forecasting finalcosts, butdo tend to show the outcome if no
improvement is made or action is taken. In essence they
challenge 'Show why we should not take these as the likely
outcome'.
This form of analysis can also be shown graphically using
S charts, both for the total project and for individual cost
accounts. Figure 8.3 shows an S chart for a typical project.
The cumulative data elements are plotted as shown, and
both the cost and schedule variances are clearly shown. In
addition it can graphically show the estimate to complete
(ETC) and estimate at completion (EAC).Figure 8.4 shows a
similar analysis for a cost account in the simple example.
Cost variance and schedule variance when considered
together, give an integrated picture of the position of the
project, or individual cost account, as shown below.
Cost Scheduled
BCWS BCWPACWPvariance variance Analysis
-1
-1
On schedule,
on cost
On schedule,
under cost
On schedule,
over cost
Ahead of
schedule, on
cost
Ahead of
schedule, under
cost
Ahead of
schedule, over
cost
254
Cost Scheduled
BCWS BCWPACWP variance variance Analysis
5
-1
-1
-5
-1
-1
Behind schedule,
on cost
Behind schedule,
under cost
Behind schedule,
over cost
Figures 8.5 and 8.6 show the performance analysis reports

for periods 2 and 3 for the simple project shown in Figure 5.5.
To illustrate the terms used, consider the cumulative performance analysis for cost account 2 in period 3 as shown in
Figure 8.6.
Cost variance analysis
The comparison of BCWP and ACWP shows whether

completed work has cost more or less than was budgeted for
that work, for example
Cost variance = BCWP - ACWP
= 7650 - 8370
= -720
Thus the cost account is costing more than budgeted.
Analysis of the differences should reveal the factors contributing to the variances, such as poor initial estimate for
the task, technical difficulties requiring additional resources,
the cost of labour and materials different from planned, or a
combination of these or other factors.
Schedule variance analysis
Comparisons of BCWS with BCWP relate work completed

to work scheduled during a given period of time, for example
Scheduled variance (in terms of cost)= BCWP - BCWS
= 7650 - 9000
= -1350
Thus this cost account is significantly behind schedule.
Figure 8.4 6 curve for cost account 3

256
PRRFO?!:~:~CI:
COST A C .
:!UI1SXP.
COST at.
I!U:IPEE
EIIDGGTE!? c ~ S ?
!IODf SCI!. TIORT: P C .
? ~ n w ? r n COR!70P7C SCl'.
r!hlr'
7%
AI'PLYSIS PO!? P e n I o D 2
I:~c::
RCT.CO.S~
!!Oi?K
PE.
VA?I.A:ICE
SC!IEDULt:
COST
PZ??.
SC!ICIIULE
nc- .cnxr.
VAP.IA:IC".
SC!!RlllbS C O P T
PERF. 11-sx
scPEnu!.s
COST
!'O*!:
?C.
Figure 8.5 Performance report for period 2
COST
Gl!FO*'?
EAC
'IAnTn'~CE
PEEP.@R!!A!!CC:
COST AC.
l!lIMTIER
BUDGBTSD COST
~ 1 Q rSCII.
.~
WORK ?Z.
1
2
3
1
C
51d?.8C
5000
0971.42
(?
COST hC.
NlJ?lBER
D W G E T E O C05"
"Oq"
.. SCH. W R K PC.
0
(410
47CC
2400
hAALYSIS BOP. YERIOD 3

SCHEnIlI,r COST
5100
-73?.Rf
- 300
-2571.42
6000
3500
~TI!!~L.?~IVI:
?rwFbn''?''??
I!T.~F,"SIS
wnn:: mc::r.cr
WP
1IIDCET
?.~l.r,':i.
ACYOAI.
.ICY. COZY
WORK F?.
?C?f JILIOCX
S C R Z 3 l J L P CO.?T
'1ARI.ANCE
AC1.COST
k'Cr?K PC.
-1390
-1?AO
Is
'!A?II:,NrS
Figure 8.6 Performance report for period 3
DCYTO-
rI&RI;hlCC
SCI!B9llLZ
@
F.P57502
C.?C
o.dC275o
0
-GPO
rOST
d
0.46$705
0.7~3333
P.677356
n
PERF. INPS::
COST
SCHEOnLS
B!lOC,?T
EAC
~!~RldSCe
260
Although this provides a valuable indication of schedule

status in terms of monetary value of work accomplished, it
may not clearly indicate whether or not schedule milestones
are being met, since some work may have been performed
out of its planned sequence. A formal time scheduling
system must therefore be used to provide the means of determining the status of s~ecificactivities or milestones, as
describued in the chapte; on planning. For example, Figure
8.7 shows the integration of the cost and schedule analysis at
the end of the thGd month in this example.
Pet$ormance indices
On quickly scanning such a report, varying figures for cost

and scheduled variance may at first glance not show up significant deviations clearly enough because of differences in
magnitude of the figures involved. For example, a cost
variance of 20,000 on one cost account with a BCWP of
200,000 is not as important as the cost variance of 20,000
on a cost account with a BCWP of 100,000. The use of cost
scheduled performance indices overcomes this problem
and provides at a glance a reliable indication of danger
spots.
Cost performance index (CPI) = BCWP + ACWP
Schedule performance index (SPI)= BCWP + BCWS
A performance index of 1 then represents 'par' performance.
An index performance of less than 1 represents performance
poorer than planned; an index greater than 1 represents a
better performance than planned. Thus in the above examples, the first cost account would have a cost performance
index of 0.91 and the second cost account would have a cost
performance index of 0.83, clearly a greater problem. In the
case of cost account 2 in Figure 8.6.
CPI
765018270
= 0.914
SPI
=
=
7650/9000
0-85
These variances and performance indices are required atthe
LA
PERFO
NCE ANALYSIS
261
levels. Performance analysis

cost account level and at
of direct summaries of
of higher level cost
basic level. It is thus
the results of such
relatively simple to deterhnine project status and organisational performance at all heve~sof both the work breakdown
structure and organisatiohal structure from the cost account
level. Cost favourable va$ances in some areas may be offset
by unfavourable variances in other areas, and higher level
managers will normall$ see only the most significant
variances at their own le+el. However, the accumulation of
many small variances which may be adding up to a large
overall cost problem, nok attributable to any single major
difficulty, will also be evident
In this simple example] the project is the only higher-level
cost account, on the vertical consolidation axis, for which
performance analysis is darried out. On the horizontal axis.
performance analysis is Arried out for the two departments
involved. Figure 8.6 s h o d this analysis carried out for these
three higher-level cost a*nts.
Forecasting the final cost

Forecasting the final cost bf the project and parts of it, that is,
cost accounts andwork b eakdown structure level items, are
pieces of information oft n requested of project managers
by anxious senior management and are vital to the project
cash flow, the viability of the project and sometimes whether
to cancel the project after it has started. No matter what the
reason, as soon as word has started, this information is
frequently and urgently dequested.
A great deal of managlement judgement should go into
these estimates, but this sgstem does provide a mechanistic
way of estimating these final costs, as there is a danger of
relying too much on subjective estimates. This analysis
provides a starting poinr for more judgemental estimates
and a benchmark to comdare them with. The method used is
a simple extrapolation ok performance to date. The forecasts are calculated for each cost account and higher-level
262
accounts, including the project, for which a full performance

analysis is carried out. They are calculated by dividing the
budget for the remaining work to be carried out by the CPI
and adding it to the actual cost of work performed, for
example
= BCWP/ACWP
Performance to date (CPI)
Budgeted cost for the remainingwork = BAC-BCWP
= (BAC-BCWP)/
Estimate to complete (ETC)
CPI
Forecast cost at completion (EAC) = ACWP-kETC
Using this method forecasts of the final cost for segments of

the project can be made for
1 Individual cost accounts that have been started.

2 Work breakdown structure items that have been
started.
3 Organisational cost accounts that have been started.
Thus, forecasts of the final cost of all activities and cost
accounts that have been started can be made, based on
performance to date. In addition, using average performance to date, forecasts for the expected total cost of the
project can also be made.
In the example
= 0.88
Performance to date (CPI)
Budgeted cost for the remaining work = 75,000-29,150
= 45,850
Estimate to complete (ETC)
= 45,850/0.88
= 52,073
Forecast cost at completion (EAC) = 52,073+33,107
= 85,180
One drawback is that at first sight no forecast can be made
for those segments not started, except to take into account
the average overall performance. This is where the hierarchical cost account structure permits a logical extrapolative forecast for many cost accounts not yet started. For
example, where an organisation is responsible for several
cost accounts, performance on their initial cost account can
263
be used to forecast performance in later cost accounts.

Admittedly there are many factors to take into account, but
the extrapolative forecast is a good starting point and any
change in it must be justified. Similarly forecasts of final cost
on higher level breakdown structure items, can be based on
performance on the early work
In this example cost account 5 is not started but the CPI
for the department involved is 085. Using this as a measure
of future performance the EAC for cost account 5 can be
forecast as &16,451.
There are thus three methods of forecasting the final
project cost, namely:
1 The average overall performance to date.

2 Organisational performance.
3 Work breakdown structure performance to date.
In this example, only two methods are used, namely average
performance to date on the overall project, and performance
by the individual organisation. This gives an EAC of85,181
and 85,931, respectively, compared to a budgeted figure of
75,000.
This form of performance analysis can be carried out
simply and quickly using the computer, with no more data
required than that which is normally available. Figure 8.8
shows the layout of a typical computing system, which
requires only three raw data files to hold the data involved,
and Figure 8.9 shows the contents of these files.
File 1 - contains the simple network data.
File 2 - contains the estimate for eachworkpackage,
the cost account to which it belongs, and the
hierarchical structure of cost accounts.
File 3 - contains the data required to produce a
performance report for each period.
The network programme module takes as input File 1
containing activity data as shown, and calculates the time
analysis and bar chart. It produces the scheduling data as a
report to management, and creates a data file to hold this
information for the remaining programming modules. In
update plan dafa
lnirial plan data
FILE I
FPH/PEBT
Frogram
Estinate and cost
etruccure data
Schedule
FILE 2
data
~ c h e d u l ereport
Budget
Proerar
~ o n t h l ycost data
Flu 3
I
Budget
dara
Budget report
Perfomnee
analysis
prosram
cumulative records
Figure 8.8 Flow chart for computer programs
264
266
more realistic cases, manpower data would be used to

produce manpower loading details and some rescheduling
carried out, before producing the final schedule. The budget
programme takes the schedule data held in the file, the
estimated cost for each work package, or activity, and the
cost account code from File 2 and produces a time-phased
budget as shown in Figure 8.1. This can be produced for each
work package, cost account. work breakdown structure item
and organisational cost account for each time period and on
a cumulative basis, that is, BCWS and cumulative BCWS.
Thereafter as actual costs and progress on work packages
are reported, the analysis programme calculates the ACWP,
BCWP, cost and schedule variances and performance
indices, and forecasts of final costs for all cost accounts and
the project as a whole as shown in Figure 8.5 and 8.6. This is
done simply and automatically involving no more work, or
probably less, than is done in manual systems.
This is a simplistic example, but no more data files are
required for each activity for a project with several thousand
activities, to produce the same type of performance analysis
with the same ease. This type of analysis can give a project
manager real control of his project, often for the first time.
Without it he is essentially sitting in a rowing boat in a river,
with no oars and no rudder. The current simply carries him
along and he and the boat may eventually reach his
objective, but he will have had little influence on how or
when he got there. In more sophisticated applications these
data files would be in one data base and a data-based
management information system used. A hypothetical
analysis for the first three periods in the simple example
might proceed as follows.
Period I. The performance report for period 1 is not shown,

as the only work done was on cost account 1. The budgeted
expenditure (BCWS) was 5,600; the actual expenditure
(ACWP) was 5,550 and the package was estimated to be 80
per cent complete. Thus, earned value was 5,600 (80 per cent
of 7,000, the total estimated cost) that is, BCWP. Work is
thus estimated to be on budget and on schedule.
267
Period 2. The performance report for period 2 is shown in

Figure 8.5, both for the individual period and on a cumulative
basis. In both, the expenditure is approximately as budgeted
and the performance indices are only marginally below par.
The forecasted final cost, that is, estimate at completion
(EAC) is only slightly more than the budget at completion
(BAC). Apparently there is no cause for alarm. In fact there
are several small signs of problems:
1 Schedule performance index (SPI) for cost account 2
is low at 0.84. This is caused by the late completion of
cost accounts which delay its start. ow ever, the
momentum of work seems to be approximately as
planned and budgeted, so that this slow start appears
to have been overcome.
2 The cumulative cost performance index (CPI) for
cost account 1 is now reliably estimated at 0.91, being
based on a completed work package. If this is used
for department A's performance, it raises the EAC
for its work to 30,010; if used for the project as a
whole it forecasts an EAC on 82,417 which is rather
more worrying.
3 The CPI in period 2 is 0.67 and this abnormally low
value is caused by the system correcting for the
previous period's optimistic estimate of percent
complete of 80 per cent, as against a more realistic
value of 72 per cent as can now be estimated with 2020 hindsight. This demonstrates how the system
rapidly corrects and highlights previous optimistic
subjective estimates of progress. It also calls into
question the reliability of the percent complete
estimates of the three work packages that are open in
this period. If they are similarly overestimated, then
the project is heading for a major over-run. However,
it is early days yet, the signs are not conclusive, and
the project manager would have difficulty in persuading other people, but at least he would be
alerted.
Period 3. The performance report for period 3 is shown in
Figure 8.6 and the problems are now more obvious. Though
- -
268
ADVANCED PROJECT MANA G EMEN T
the rate of expenditure is only slightly below budget, the

work achieved for this expenditure is much less than
expected. The significant pointers are:
1 Both the SPI and CPI for the work done during the
period are very low, indicating both that the work on
the project is not going well and that the previous
period's estimates of completion were highly optimistic.
2 The forecast final cost of the project (EAC) is now
85,131 based on overall performance, or 85,931
based on departmental performance.
3 Both departments' performances are well below par,
with department B's being especially low at 0.837 for
schedule and 0.851 for cost.
4 Work on cost account 4 is well behind schedule and
well over budget, that is an SPI of 0.7 and a CPI of
0.83.
Thus performance analysis using the earned value concept on costcentes, termed workpackages, each looked on as
a discrete subcontract, and structured into cost accounts
using the work breakdown structure, can provide effective
control of projects. The size of the work package can be
varied, according to the size of the project and the sophistication of the information system used. The schedule and
cost variances, coupled with schedule and cost performance
indices give sensitive and reliable indicators of progress
achieved against plan and budget. The system also automatically produces forecasts of final cost for parts of the
vroiect and the vroiect as a whole. The verformance of
contributing organisations is automatically^monitored and
gives an effective feedback to those involved. Hovefullv the
i s e of performance analysis will lead to more effictive
control of projects in the future than in the past.
A
<
Data overload
The presentation of the performance analysis reports shown
previously is satisfactory for the simple example used, but
269
obviously inadequate for the larger project where hundreds

of cost accounts may be involved. In such cases the project
manager is faced with one of the common problems of
project management, data overload. In larger projects there
is so much data required and produced for control that the
project manager can be faced with literally hundreds of
pages of computer output each month, which he just does
not have the time to assimilate.
There is a major difference between data andinformation,
and what may be information at one level of management,
may be data at a higher level. Thus it is necessary to have a
clear understanding of the difference between data and
information. Data can be defined as groups of non-random
symbols which represent quantities, actions, things etc.
Information is data that has been processed into a form
which is meaningful to the recipient and which is of real or
perceived value in current or prospective decisions and
actions.
Data analysis for information

Thus the project control and information system must be
able to analyse the project data to produce meaningful
information to all those involved. This implies that each
manager involved on the project receives the information
that is relevant to his responsibilities on the project, no more
and no less.
Even if data is effectively analysed to produce this
information, on the larger project this will still produce a
mass of detailed information which needs to be consolidated
and filtered or reported by exception.
If the project, all the component elements of it and all the
functions involved are performing satisfactorily, i.e. the
project is progressing to plan and budget, there is no need to
report additional detail. Similarly if only one element or
organisational function is showing a significant variance,
then only that element needs to be fully analysed and
reported. In addition each manager involved on a project
270
only requires that information relevant to his area and level

of responsibility.
Performance analysis information must be consolidated
and summarised to provide the relevant information to each
manager in the work breakdown structure and the organisation breakdown structure. It must also be filtered, i.e.
significant variances must be highlighted by exception
reporting and narrative reports made as to the reasons for
them and the corrective action taken. Troubleshooting time
and effort need then only be concentrated on problems
having a significant impact on cost and schedule.
Coding
The project control and information system thus depends
on a coding system to identify the structure of cost summarisation for both the work breakdown structure and the
organisation breakdown structure. The coding system must
also be able to handle both the cost account or work breakdown element itself, e.g. electrical drawing office, and the
cost elements involved are labour, materials etc. This permits
consolidation by work element or cost element, e.g. higherlevel reports could examine consolidated labour performance as well as overall performance on major parts of the
project.
Thus a coding system may have three reporting structure/
summarisation tables to facilitate this:
1 Work breakdown structure.

2 Organisation structure.
3 Cost element structure.
One such structure is shown in Figure 8.9.
The consolidation procedure should then be able to
summarise data and information of any of these reporting
structures, either separately or in combination. The work
breakdown structure permits reporting in the way that work
is planned and carried outwhile the organisational structure
permits reporting on functional department or organisational performance. Reports at different levels would then
27 1
give different amounts of detail from senior management to

cost account manager.
Cost account report

The basic building block of the control and information
system reports is thus the individual cost account performance report, as shown in Figure 8.10. This goes to the
manager of the cost account and shows the normal performance analysis information, as described previously.
Higher-level reports
The individual cost account reports can then be consolidated
into the reporting structure as shown in Figure 8.11. Thus the
senior management will receive the consolidated reports for
all projects the company is involved in, the client and contractor project managerwill receive the consolidated reports
for the individual project, the drawing office manager will
receive the performance report for his department's performance for all the projects going through the drawing
ofice and the individual project The manager in charge of a
WBS element will receive a performance report on that
WBS element, and so on. Individual managers will receive
reports for their span of responsibility at a level of detail
appropriate to their level of operation.
Exception reports can be produced at any level by setting
threshold levels for size of variance or deviation from the
norm in performance indices.
Traceability
Hand in hand with this process of consolidated reporting
must go the principle that variance or deviation from the
norm of performance indices must be able to be traced
downwards to their sources. Thus if a project is seen to have
problems, it must be possible to look at the next level of
I E ~ a o o o1"5 "
5
:A
12'2
C 3 I-IW I I
e,!
I l l
k w
IWg:
:
l ~ a
3 1
I
- u
-> - a \ O h
lnh0000
I*-00
U 3 . I O " O O
momcN
C
" .-
c 3
3 a
0 L
C- @ b
a 0-
0 3 lmvl
- U m L
Ye5
l o m o o
.L
*a m
2 .
--a
C Y
'
L
t.
6"
3
*
m
c a
$ 4.
U'
03
- -
IO-.OYOO
1 0 - 0 0
o*no
/C~NN
4
0 0:
.*
TJZ"3C
0
u q
I a m o o o o
10n.00
0
j o ~ u l o o
I V n N r i
0
1om
D
0
0
fan.?
=
-2
'
12
I-
*
*
I
I
...
2 +te
m
~m I
a
ID. I
0
~i.
I a
1m 1
e
11
I-.
I
I
0
I
0 lo.
1
o lrn I
10- I
I
1
0 . I N . ~ O O o
0 3 l i n O ~ 7
l"tY
10
I1-
I
~ O O O O C 1 0 10 1
~ U I I O U O O
om:
%
=
U ti
'z
"
* r
(1
I
I
I
I
16 I
? r.
O W L
o lrn 1
1'4 1
0 l m I
5:
I
II
1000000
UO I - N O O
.-r i
I"
1"
o IN
N lm
I N 1
a.0 J
D L
U-f
...-..
...
??I!
0
0
..8;- / R $ T ~
* . - C
:1!; 11
I
2 -6
0 - 0
.+
- -
1-1
I:
1
Irn
I
I
h
0
o1
0
1
LL
i
I
I
I
I
I
,
I
I
I
I
I
L.
8
L
C)
*
ul
I
0
,
I
a
M
iE
272
2
':
,. j,z . B.
.:
?
=
'
- i;.
2"
I:
.:
;;"
;; j
,
::
, . .,
1'
<
..
.m-
- -:
....
-m.-
"ac.
;-x=
-:. ;i
2 -.& :
a;:
1.;;
f &;
2 ;
-
, ,
I:
.. !
.,. ..*
.c iZ
:
jr
g-;;
3 I'
".
:
.-<
"
. i
"rl
;"5
-3
:'z
i3
Y Y
"
".
" .
$2 /
: ,;
<!
'
? :
:%;
2, i
-2; /
5; :
=:
-> m5 <
m :-
a;;;
:j
1 8
* . , ,
4.4:
I?!
5
"
"'I
: ;. i'
g, ; ....
""
&,,a*
.c
. , - ,
"?%
:i
,a,.rz
.* - c
.<........
5 "
:i
:f
e.
3;:
;E:;
:"a:
- - -..-.
.o-.-h
: ~-
$2;;
. -..
:?;:::E
'-"==-s
11:2::2:
. . .:;::;.:: . . . . . ., 2
:f
: b
;:!::
;i /
.!;Pi::
:
'
............ ! 5' . .,-:
,
..
- e s m .<=.
-.Lo
:;ii
.<-
.;
u-X'
E " .
,.&
0 . .
..,.,.:y..-
# ,
-".&
t; ,;I
21
<-..
u8jiJ
2;
. ....
. -..
.
.
.
:
:
z
ia.,
--.-.
.
.,
...
.
.
.
.
.
.
...
-.........._ .....$.'Z'
-....
...............
I
E
: L
-.-: .
::;:
--
111
,,
.-.
:
-.
:1
'":
5:' '1
:I:
., L1 :,, ,:
-.
"6
U L
." ,;, ->o z - s.a.i

; ;"
,. >;:
-*:.
a .
:."
<:
i
5 :=:
&;
3 8:
I;,
5.
2 ii
"-
., ->a.
m;"
";
;
. - --Y
.
B.;' [ " >5 ..
2:
3. ;i
;<z,
z ;
;4 jj
-.
:I
.1
= #
::.r,
,
;
:I
.:
6
-.
m.
,
;:a.
:*
,o
: = I, "- 7,;
,=
-E
,,
;-,I;-
::*:
::
,
:
-.a
,,.
,. -::e- =
0 d"
, b
-*
....;
;
.F
"&
,:
.:,;:
.."
.
,.
ij
2
2.
,,
.
:
!-
;;
zz
="
/:
A>
w *
:*
, a
<8
i : i /,.; : :....
;
...,
-=.
-.- ;
i
, -,
a ii"i"'2
,
.:z
.-
:3; , a
:.#,; c : >
2
:
" *,
;
3; 2
- 2 ij
-3:
{t
,-
a:
..
.,
-u
!y I ;
~.
:a i;,
.
- ;;-
.-
,, -- ,
. . . .u .,, .. . .
t, ,; i :
.-- -
--
z -
:n
G:;:
:, -- -*-
*
.
, %
,, .--
,.
,-
1:
. I
.r . 1 Y 1
Y I
-".*I
r o r l o
01
. I 0
" I
- I
* I
C . - C I - I
r , - 3 - 1
3:s:
O :
OI L O C I
."
C ' O ' ? '
_ I
0
1
0
1 - 1
D C O I C I
?!
1 - 1
I
I
; a r k : ~ l
l&l;ii:l,!I
7--T---
275
breakdown, to see which level 1 element is causing it. This

process must be able to be continued to cost account level
and the cost elements involved, and then if functional
performance is seen to be in question, it should be possible
to trace up the organisational breakdown structure to
examine performance on other cost accounts and higher
level elements. A simple example is shown in Figure 8.12.
If interactive computing systems are employed it is quite
possible to sit at a visual display unit and cany out the
tracing operation quickly and easily. However, no matter
what system is used, this traceability is inherent in the
coding structure used, together with the work breakdown
structure integrated with the organisational breakdown
structure. In the example shown in Figure 8.12, the project
has a schedule performance index of 0.89 and a cost
performance index of 0.95. If it is chosen to trace down the
source of the deviation on the CPI:
1 The project report shows the problems are in basic
design and the electrical sub-assembly. If it is chosen
to trace down the electrical sub-assembly structure:
2 The next level shows the problem is confined to the
electrical work element.
3 The next level shows the problem is in the electrical
design and electrical drawing office. If it is then
possible examine the electrical drawing office cost
account report, as shown in Figure 8.10.
4 It is then possible to trace up the organisation breakdown structure and examine performance on all
drawing office work This shows that schedule performance on the electrical drawing office cost
account has been completed at the expense of
schedule performance of mechanical and instrument work. It also shows a major cost performance
problem on the mechanical drawing office which
can be traced up its structure.
Control of changes to a project

One of the identifiable common causes of delay, cost
276
escalation and low labour productivity are changes to the

project. Many, many projects take longer and cost more than
necessary because of changes and additions to the original
scope. It is quite common for changes to a project to add
anything up to 50 per cent to the project cost and not unusual
for them to add more. Thus one of the most important and,
unfortunately, unpleasant and troublesome functions of a
project manager is to control the changes to the project.
Changes have the following detrimental effect on performance in project work, namely:
1 They increase the cost.
2 They cause delay.
3 They reduce morale and productivity.
4 They worsen relationships among those involved.
Changes in design in project work normally lead to additional cost and time to completion. In addition to this extra
cost and time, which may or may not be justified, design
changes during the engineering and construction phases
cost more and the work takes longer than the same work
would have if it had been included in the original project
specification. This is because, in addition to work having to
be redone, the sequence of work is disrupted; not only must
the design of the particular item concerned be changed, but
all those items interrelated to it be changed too. At the construction stage, changes are more costly still, because work,
materials and equipment may have to be scrapped and
design, materials purchasing and construction work must
often also be done on a rush basis.
Though changes have a detrimental effect on morale and
efficiency in general, this effect is multiplied in the manufacturing and construction phases. Wheneverw~rk
has to be
torn down or scrapped and repeated, labour morale and
productivity is affected and this can lead to long-term effects
not the least in the area of industrial relations.
One of the worst effects of changes is on relationships
between the project manager or managers, and other
managers on the project. It is the project manager's,responsibility to resist changes unless they are absollCtely
essential to the objectives of the projects and in the
277
company's interests. This forces him into conflict in two

ways. If the project manager is in the owner company, he
must resist changes arising from his own company's
managers, and also negotiate changes with any contractors
involved. Changes, as mentioned previously, disrupt the
design and construction work and are more costly than just
the work directly involved. Thus the actual cost of changes is
much higher than the workitselfwould make them, and they
are a very difficult thing to estimate accurately and to negotiate. In addition, there are often legitimate differences of
opinion as to what is actually a change to the contract, or
what the cause and effect are. Some contractors also tend to
overcharge for extras, and make a significant percentage of
their profits in this way. It is therefore often very difficult to
disentangle these factors, and this uncertainty can cause
differences of opinion and hence conflict. Probably the
worst of these effects is on the -proiect
- manager's relationship with other managers in his own compan<be it client or
contractor. The differences of opinion and conflict that
occur have often to be resolved a t a very senior level in the
companies involved.
Changes in a project can also lead to a breakdown of the
control system. Sometimes actual progress, resource usage
and cost are compared to planned values of these factors
which have not taken into account changes leading to a
totally misleading assessment of performance.
Changes may occur at any stage during the project and
can arise for several reasons, namely:
I Changes in the specifications or scope of the project

during the development stages. Changes to design
occur legitimately in these stages, but often complex
changesin scope or specification are proposedand
accepted implicitly without adequate evaluation of
the consequences in terms of time and money. Once
the project specification and cost have been approved, changes in scope are very expensive and
have a questionable value.
2 Late design changes are the most troublesome and
expensive changes of all. They normally arise out of
errors and omissions in the original design work,

afterthoughts, and from the client company attempting to keep up to date and incorporating the latest
developments in technology. There is little alternative in incorporating changes to correct errors and
omissions, but afterthoughts and changes to keep up
to date are a bone of contention in project work
3 Changes due to safety or legal requirements. Here
again there is little alternative but to include these
changes in the project.
4 Changes which are justified or appear to be justified
to improve the rate of return. The advisability of
these kinds of changes is debatable. The fact that a
certain addition to a project is estimated to give a 30
per cent internal rate of return may not be sufficient
justification to include it. Whether or not to accept
this type of change should be a matter of top
company policy. One view is that they are not
acceptable because this type of forecast rate of return
is often inaccurate and biased; estimates of the cost
of changes are very difficult to make, as discussed
previously, and thus the rate of return is also difficult
to estimate.
More importantly, if the project is, say, to build an
oil refinery at a certain capacity, changes of this kind
do not usually add any overall significant advantage
on a company or market basis, and the objective
should always be to carry out the company's top level
objectives as cheaply as possible, consistent with
satisfactory time and technical performance. Many
worthwhile projects have been failures because of
additions of this nature. If spare capacity is worthwhile, this should be a top level management decision and not some design engineer adding a slightly
larger size of equipment.
5 Changes which are thought desirable. This is a real
area of conflict, particularly within companies.
Operating managers would like certain additions for
various reasons, sometimes worthwhile, sometimes
simply luxuries; design engineers would like to add
279
certain features, upgrade others or increase some

equipment sizes. The project manager has to resist
this type of change and he must establish a clear
definition between 'musts' and 'wants', and only
accept changes that must be made to meet the
original scope and safety standards, that is, 'musts'.
In order to control change, and reduce the conflict within
and between companies, the project manager should endeavour to achieve the following:
1 Implicit backing up by senior management of his

efforts to resist non-essential changes. The policies
on changes to increase the rate of return on projects
should be spelt out by senior management. If this is
permitted, the rate of return used should be significantly higher than that used as a cut-off point for
normal company projects.As far as possible a policy
should be established and communicated by senior
management on musts and wants.
In other words, senior management must back up
the project manager by saying 'no' to changes which
are desirable, but not essential.
2 The original scope and specifications of the project
should be as clear as possible. Factors like the policy
of whether or not to allow for some spare capacity to
allow for expansion should be spelled out by senior
management.
3 At some clearly defined stage, no further changes
must be permitted, excepting those that are absolutely
essential for the success or safety of a project; that is,
the project mustbe'frozen' at some stage.The project
manager would like this to be frozen at the development stage, but other managers would prefer it to be
frozen at the construction or manufacturing stage.
The earlier it is frozen, the lower the cost and the
delay to the project caused by changes.
4 Finally, a change control system must be set up.
280
Change
contra[ systems
Thus the control of changes to a project is one of the most

critical activities of a project manager.Although the number
and scale of changes to a project will vary with the stage of
project specification, e.g. between a fully specified fixedprice contract and a research, development, design and
manufacturing project, the importance of the control of
chanee does not. One would think that chanee would be verv
limitGd on a fixed-price, fullydefined projectwith completed
drawings and perhaps bills of quantities, but also in practice,
this is not the case.
However, it is in the larger project that the control, or in
reality the management, of change really reaches a climax.
Although change must be resisted, it is generally impossible
to avoid it in the larger project. It must be accepted at the outset of a larger project that change is going to occur and
change control systems must be set up to manage and
control change as a way of life.
In order to identify, analyse and control changes to a
project, the project manager must set up a system to handle
these changes. This system must:
1 Identify changes from the original scope.
2 Forecast their cost and their effect on otherwork and
the time required for them.
3 Subject them to managerial analysis and decision
4 Record actual figures on them.
5 Highlight them to senior management
6 Set up a system for solving disputes with the minimum
of conflict
7 See that the changes are implemented.
The implementation of a change control system, sometimes
termed 'trend forecasting', 'deviation control' or 'configuration control', early in the project life is critical. Such systems
must emphasise periodic weekly or monthly reviews during
the formative design and procurement stages of a project.
Control is made effective during these vhases bv frequent
reporting of changes and by thecomm&ication-of change
notices to key people, so that full consideration can be given
281
to the necessity of the requested changes and their effect on

cost and time.
This system must be a dynamic technique of estimate and
specification revision, by which the effect of departures from
the basis of the estimate and specifications are identified as
they occur, and adjustments are made to the control estimate
being used at that particular stage of the project. This
enables those in control of the project not only to see the
effect of change on schedules and cost much more quickly
than previous techniques allowed, but also it enables them
to consider corrective action, and in many cases to take such
action at a time when alternative courses of action are still
feasible to counter the effects of adverse variations. Thus a
change control system must be considered to be a system of
continually identifying changes, or deviations as they occur,
from the conditions or facts assumed in the project estimate
and specification baselines and of taking corrective action
where applicable to counter the effect of adverse change on
the cost or schedule.
Change control systems have evolved in different ways
with different jargon in industrial projects and in advanced
technology industries, primarily the aerospace weapons
industry. In this industry change control is called'configuration control' and is just one part of what is termed 'configuration management'. Although this is an esoteric term
for construction projects, it is widely used in advanced
technology projects and implies a formalised discipline for
providing uniform product descriptions, status records and
reports. Configuration management is essentially the
formalisation of a discipline which integrates the technical
and administrative actions of identifying, documenting,
controlling and reporting the functional and physical
characteristics of a product or project during its life cycle
and of controlling changes to these characteristics. Thus
configuration management is a more embracing term than
change control system and has more emphasis on the initial
structuring of a project and on an engineering change
control system to ensure the product produced is the product
desired. However, configuration control and change control
systems are more or less identical processes involving the
following steps:
282
1 Establishing a baseline scope, specification, estimate, and schedule for the project.
2 Identifyingchanges to this baseline, communicating
them to all concerned and determining their consequences.
3 Reviewing, approving or rejecting these changes.
4 Implementing these changes.
Establishing a baseline
A change control system should be implemented as soon as
the initial scope, specification and estimate of cost has been
defined. It does not matter how preliminary this scope and
estimate are. thev must be used as a basis for identifvine
changes, evalua6ng their consequences and controilini
them. As more information becomes available.. as ~reviouslv
described, the project scope, specifications, plan and estimate will become more refined and at each stage will form a
baseline for the future control of changes.
The change control system should enable management to
trace the reasons for the differences in estimates made at
varying stages in the project. Often the question is asked by a
client or contractor's management6Howdid a&10mestimate
at an earlier stage come to b e a &15mestimate at this stage?'
As described in the chapter on estimating, the quality of an
estimate depends to a large extent on the amount of
engineering that forms the basis for that estimate. As
engineering progresses, better quality estimates can be
made. A change control system bridges the time gap between
a lower quality of estimate, based on less well-defined
engineering, and the higher quality of estimate that results
from more detailed engineering. Typically on important
projects. a design specification is written at a relatively early
stage. to define the particular design that has to be used in
the project. An estimate is made of the cost of this plant and
approvals are sought and obtained. Let us assume that this
point in time is point A. Additional engineering must be put
into the project. The detailed process design is developed
and mechanical specifications are completed to the point
where a fixed price competitive bid can be requested. A new
A
283
estimate, on the basis of this greater engineering definition,

is made at point B in time.
A-B thus represents the time interval during which the
detailed design was carried out and could represent anything from several months to several years. Unless there is a
good estimating technique, combined with a change control
system, the project could easily get out of control. The
change control system prevents this happening and makes it
possible to trace the reasons for the differences in the two
estimates.
Ident13ing changes
Before changes to the project can be controlled they first

must be recognised. A change or deviation is identified as a
noticeable departure from the control estimate, specification
or schedule. It may be favourable or unfavourab1e.A change
occurs when any factor including design, but not limited to
design, varies appreciably from the control. Thus a change
in process sequence or addition of an extra pump or deletion
of a pump, or anincrease in the price of materials, or delay in
construction, all represent changes. Similarly, a more complex design than that estimated for, represents a change.
Recognition of deviations can best be made by those
parties directly connected with the details of the project. An
estimator can notice a change from the assumptions he
originally made. A design engineer knows when the scope is
changed, or when he has to add equipment not originally felt
necessary. The project manager is not only constantly aware
of pressure for change, but should be aware also of general
price increases. The important point is that everyone must
conscientiously be on the lookout for deviations to the
project and that it must be a co-operative endeavour since no
one person can identify all the changes that can occur. Thus
it is the responsibility of everyone involved in the project to
bring to the project management's attention any change
from the accepted baseline of the project. This must be done
by the individual who first becomes aware of any change, as
soon as he becomes aware of it, by making up a change
request form and submitting it to a change control group
CHANGE REQUEST
.................
...............
...................
Number:
Revision:
PROJECT:
ITEM AFFECTED:
Date:
Nam:
Work package:
Item No($):
C h a n g e requested by:
DESCRIPTION OF CHANGE:
REASON FOR CHANGE:
Code:
1TEMSlAREAS AFFECTED BY CHANGE
FIRM ESTIMATE:
I N I T I A L ESTIMATE OF COST/SAVING*
EFFECT ON SCHEDULE:
CHANGE APPROVED/REJECTED*
- CONTPACTOR'S
SIGNATLIE:
..................
D A Z:
C L I E N T ' S SIGNATURE:
*delete one
Figure 8.13 Change request form
DATE :
..................
285
meeting for approval and further action. One example of

such a change request form is shown in Figure 8.13. This
form:
identifies the change, describes it and gives the cost
account, work package or cost control number it
affects
gives the reason for the change
identifies who is initiating or requesting the change and
provides for his signature
gives a first descriptive appreciation of the consequences of the change and the segments of the
project it affects
gives a rough estimate of the effect on the project
schedule
gives an order of cost estimate of its effect on cost
can optionally give a code number to identify the cause
of the change for post-project analysis, for example,
customer request, late design change, error/omission
in design, legal, environmental reasons, increase in
economic return.
Additionally, it provides space for a firmer estimate of cost
and time, and a record of approval or rejection of the change.
Change control board meeting
All the above change request forms would be submitted to a

change control board chaired by the project manager and
made up of key managers on the project This should meet
frequently, at least once a week in the formative stages of the
project, and be a limited attendance 'stand-up' meeting
concerned more with communicating changes rather than
problem solving.
At this meeting, which should not take more than half an
hour, those parties working directly on the design, estimating and control, table anticipated and potential changes
known to them, by using the change request form, that have
not previously been tabled, or that they have changed since
last being tabled.
Subsequent to this meeting the cost control or estimating
286
representative assesses the guestimated costs, properly estimates the changes having a significant effect, allocates a
serial number to the change and Dasses the form to the
project manager. He also prGares a summary report showing
the effect on the estimate of all vrobable and potential
changes to that date. The base estimate as revisid by the
changes can be considered to be the current most probable
cost of the project.
Thereafter it is the project manager's responsibility to
ensure that all groups affected by the change know about it
More importantly, he must review the change in conjunction with only the relevant people concerned and come to a
joint decision on the acceptance or rejection of the change,
or send it back for consideration of other alternatives. The
project monthly report should then include a list of changes
arising in that month with their probable cost, and a
summary of changes to date on the project
Such a system has been found to have the following
benefits.
It provides a realistic and current most probable cost
and schedule for the project and thus avoids periodic
drastically changed forecasts.
It identifies potential changes of questionable value,
which in turn can be eliminated by good control procedures.
It identifies areas where corrective action should be
initiated to counter adverse changes.
It establishes a disciplined approach to change in cost
factors.
It makes all personnel more aware of the effect of
change and its economic impact on the project
It provides a means of subsequently analysing all
changes as to cause and effect, so that future projects
can be better planned and estimated.
It stops the acceptance of changes requested verbally
and requires those requesting a change to make it legal
and official by signing for i t If they are reluctant to do
so then the need for the change is highly questionable.
This removes a common cause of conflict in project
287
work, as very often over the long life of a project, even

with no intention of deception, memories fade and
people forget about changes they requested informally.
Integration of change control system with other project systems
The change control system must be integrated with the other

project systems as changes will affect the project plan,
estimate, manpower plan and budget. The effect of every
change must be reflected on the project baselines for control.
It must be re-emphasised that it is useless and positively
damaging to effective control to be comparing actual progress, resource usage and cost to the planned value if these
planned values have not taken into account changes that
have occurred. Thus the change control system must be
integrated with the control and information system, at the
very least, at the cost account level, e.g. as shown in Figure
8.14. The normal consolidation process will then keep the
higher-level work breakdown and organisational breakdown elements, plans and budgets up to date.
Figure 8.14 Integration of changes into the control and

information system
288
9 Project Management
Computer-based Systems
At first sight the methods of planning, budgeting, analysis
and control outlined in these chapters may look too sophisticated and require too much work to be of any practical use in
project work Many project managers may comment that
they do not have the time or the people to carry out these
methods, or that they are not appropriate to their size of
project, cannot get the kind of information required, or that
future events are too uncertain to justify the effort involved.
These comments are unjustified in that these methods
require no more data than is available for any project, and
the effort required of the project manager, or his staff,can be
less than that required for the more traditional and less
effective methods of planning and control. However, the
management of projects is a highly-skilled professional
branch of management, and the methods used have to be
sophisticated to a certain extent, if projects are to be
managed efficiently.
One of the secrets of efficient management, planning and
control of projects is an effective integrated project management information system based on: the use of the work
breakdown structure to structure the project and its systems;
the use of the work package concept to ensure objectivity in
reporting; the use of forward-looking performance analysis,
as well as variance analysis where appropriate; and the
systematic processing of the data available, whether done
manually or by computer.
All project managjrs must be backed up by effective
project systems, as it is impossible to manage a complex
entity, such as a project, without planning, budgeting,
290
analysis and control systems to assist the manager in his

function. If he does not have this back up, his planning and
budgeting will be too slow, and they will not be used to help
him organise and control his work.
Without a good information system he cannot really
control his project, as he will not know until too late what is
happening, must spend a lot of his most valuable and
scarcest resource, his time, on simple supervision and the
collection of information, and any information he does get
will inevitably be out of date.
All project managers have back-up systems of one kind or
another, ranging from the most elementary manual systems
to extremely sophisticated computer-based systems. Some
of both types are excellent but others are not worth the effort
put into them and sometimes hinder or confuse, instead of
helping the project manager. Often the simplest systems are
too slow, do not carry out an adequate analysis, even though
the raw data is there, and are not organised as an integral
project management information system. In other cases,
sophisticated computer-based systems produce piles and
piles of data print-out and very little information in a
digestible form for the project manager. The computer is
used as a high speed typewriter simply to reproduce large
amounts of data, or to give a veneer of respectability to
unreliable data. This distinction between raw data and the
information extracted from if is critical in project work
There may be from hundreds to tens of thousands of
activities involved in a large project with a tremendous
amount of data produced. Unless this data is analysed and
filtered in a systematic manner to produce meaningful
information for the project manager in a summarised and
exception basis, it will snow him under and land up in the
wastepaper basket.
It is essential that the speed of the project manager's
planning and budgeting systems must be fast enough to
manage the pace of project work, both in the initial stages of
the project and in the periodic updating required. In the
initial stages, plans and budgets must be established quickly,
otherwise decisions will be made and resources allocated
without the logical guidance provided by planning and
budgeting, and there will be no effective baselines on which

to evaluate progress and cost. They must also be able to be
used quickly, when changes or deviations occur, to evaluate
alternative ways of carrying out the work and keep the
control baselines current and up to date.
It must be possible for variance and performance analysis,
as described, to be carried out quickly as time is important to
allow management to correct deviations.It is more important
to have approximate information quickly, than accurate
information when it is too late to do anything about it. The
project manager must have information in a summarised
form and on an exception basis, on slippages to schedule,
variances from budget, delays in procurement and escalation of costs etc. or he will not be able to manage and control
his project Important points should be highlighted automatically in a digestible form. Once a deviation is identified
he must be able to trace it to its source and produce further
information right down to the relevant raw data.
The systems used will vary with the size, complexity and
value of the project, but they must always provide these basic
essentials. It is not uncommon for project managers to
become involved in a 'systems audit' of their own firm's, the
clients' or contractors' systems. This is carried to an extreme
in the US Department of Defense, CostISchedule Control
Systems Criteria standards where contractors' systems are
evaluated according to thirty-five criteria.
These systems can be manual orcomputer-based. Manual
systems can work effectively with the smaller project, if the
system is structured as described in the previous example,
but as the size of the project increases manually-based
systems become ineffective. The manually-based collection
and processing of data, and its analysis to extract information and planning, budgeting and control is tedious, involves
considerable manual work, is time consuming, slow and
rarely up to date, very inflexible and often inaccurate. It is
necessary to use what seem like sophisticated methods to
plan and control the larger project and there is a great deal of
analysis of data required if the manager is to maintain
control. If this has to be done manually, on all but the
292
smaller project, it will simply not be done and effective

control of the project work will be lost.
Nowadays it is illogical not to use a computer if it can be of
benefit. It allows us to do things that were perhaps not
possible before, but much more likely to do things faster and
more accurately, particularly if they have to be repeated
several times, with far less effort than is required with
manual systems. Normally it carries out functions which
can be carried out manually but which take considerable
time and labour and thus are not done or are not done
adequately. With a computer, integrated project management information systems become reality for even the
smaller firm.
Thus although the emphasis in the past has been on
computer-based planning, using CPMIPERT, today and
increasingly in the future, the emphasis is on a total integrated project management information system (IPMIS), as
shown in Figure 9.1. This can be made of separate modules
linked together, but more logically can be built by combining
a network package and a data-base management system.
Essential minimum system
The essential minimum of such a system contains the
following modules: planning; budgeting; performance
analysis, and change control.
Data capture
One of the problems encountered is the amount of data

input required to monitor performance, i.e. data capture of
such things as labour costs, material costs etc. Therefore a
logical extension of the essential minimum IPMIS is to link
it with the firm's normal payroll and accounting systems.
Estimating
Computer-based estimating not only leads to faster, more

accurate and systematic estim~tingbut, as the estimate is the
Estimating
Payroll
tender analysis
accounting
BOQ analysis
measurement
Materials
management 4
Planning
Budgeting
Performance a n a l y s i s
I
Graphics
analysis
Figure 9.1 Integrated project management information system
Risk
Job
analysis
canis
294
essential data base for planning and budgeting, it provides

the real starting point for planning and budgeting. All too
often work plans, manpower planning and even budgeting
are carried out independently of the estimate because the
estimate is structured differently from the work plan, e.g.
with a bill of quantities. It is essential that the same assumptions used in the estimate are also used in plans and budgets
and thus this extension of IPMIS often contains a tender
analysis module to transform the estimate data into planning
and budget data to ensure consistency.
Materials management
The planning and control of procurement is so important in
project management that a further logical extension is to
integrate the materials management system with the planning system.
Records management
Keeping track of drawings, changes and documents is such a
tremendous task in the larger projects, that configuration
management or records management can also benefit from
computerisation and linking into the IPMIS.
Ancillary systems
The following ancillary systems are also now linked into the
total IPMIS:
I Graphics - easy to use graphics packages are now

commonly used to
(a) produce graphs, histograms and other visual
reports, and
(b) produce bar charts and arrow diagrams.
2 Spreadsheet analysis - spreadsheet analysis has now
almost replaced the hand calculator in project
management and is used both separately and linked
to the IPMIS.
3 Job card systems - details of actual tasks and activities can now be produced on job cards detailing
times, resources and budgets, can also be used for
data input when completed.
4 Risk analysis - probabilistic evaluation of projects on
a risk analysis and what if simulation, rather than a
PERT basis, can now be integrated with the IPMIS.
Even a simple Apple I1 personal microcomputer system,
costing between 2,000 and 2,500 including printer and
software, can provide the essential minimum IPMIS. This
basic system can handle between 500 and 850 activities,
depending on the amount of associated data, produce
graphic reports, plot arrow diagrams (with a low cost plotter)
and implement library modules. Figure 9.2 shows the
program structure diagram for one such system, i.e. microplanner from Micro Planning Software Ltd.
If a firm is handling projects involving more than one to
two hundred activities it should really use a computer for
planning and control. In fact it can be logically argued that if
projects are large enough or complex enough to justify the
use of CPM/PERT network methods for planning, then they
are large enough to use a modem low cost, easy to use
computer system.
Figure 8.8 shows one such minimum integrated project
management information system mounted on an %bit
microcomputer. There are three modules in this system:
1 A network, or critical path analysis programme.

2 A bar chart, including total float.
3 Manpower table and histogram.
These can be produced for the project as a whole, for any
specific trade and responsibility code. Then either automatic or interactive resource levelling can be carried out to
match the plan to the manpower and equipment restraints.
This can then be modified interactively as desired to establish schedule starts and finishes to those activities which are
not restrained by other factors to give a definitive plan or
schedule for the project, both as the printed report and as a
data file for further processing.
Figure 9.2 Microplanner program structure

296
For each plan update, information as to the actual starts

and finishes of activities can be input, and if required the
above process repeated to produce anupdated plan. Options
are available to produce a free float consumption report, a
milestone report and manpower S charts.
The schedule data file and a file containing the estimate
data and cost account structure can then be input to the
budget program to produce a cash budget and a manpower
budget for activites, workpackages, basic cost accounts and
the hierarchical cost accounts as desired. This gives printed
reports and a budget data file for further processing in the
performance analysis program.
At each reporting period data and cost to date together
with the budget and schedule files are used to produce
variance and performance analysis reports as described
previously. Very little effort and no more data are required
than with traditional methods to use this integrated project
management information system, which can effectively
enhance real time planning and control of projects at a cost
most firms can afford.
Types of system
A major factor in the success of the new systems is the type of

computing system used for computer-based project planning, namely interactive real time computing. There are
three types of interactive computing systems used in project
work namely:
1 The company's main data processing computer.
2 A bureaux-based system.
3 A computer dedicated to the project management
function.
Company computer
The cheapest way of providing computer-based project
management systems is to use the company's data processing
298
computer. However, it must be a n interactive computing

system with ample capacity so that there are no limitations
on the use of the system by the project staff. Then with a
terminal in the project office they can have access to the
project management system for as long as they want, and at
any time they want it. Most computer manufacturers have
program packages for critical path analysis, though sometimes they are expensive and not up to modern standards of
flexibility, which provide a basis for building the project
management systems. Though there may be internal charges
made for the work, these are normally nominal bookkeeping charges and no 'real' money is involved. However, if
the system is not interactive and there is not adequate
capacity available, then the company's computer does not
provide a basis for effective project management systems
and some other alternative should be used.
Computer bureaux systems
Most computer bureaux have critical path analysis program

packages which are adequate for project planning, though
their sophistication varies from bureau to bureau. They also
have high-level programming languages such as BASIC
FORTRAN and APL, together with other program packages for such things as financial modelling and data base
management, with which a n effective project management
information system can be built
These systems can be operated from the bureau office, but
the more normal situation is for the user to hire, or purchase
a computer terminal (approximately 1,000) and access the
computer bureaux through the normal dial up telephone
line. This has the advantage of giving the user access to prewritten program packages mounted on a large powerful
computer. They provide a quick and easy way of getting
started in using computer systems to back up the project
management activity.
Unfortunately, with large-scale use the costs mount up,
and often people with access to the bureaux system do not
use it as much as they should. Every time they access it, there
is an indentifiable charge to the company and though the
cost may normally be well worth the service, it is visible,

often management start to quibble about it and the use
dec1ines.A more logical approach is to determine how much
adequate access to the system would cost, and to see if this
would justify an alternative way of supplying this service.
A dedicated project management systems computer

In the last few years there has been a growth of computers
dedicated to pkject management information systeks, and
with the recent develovments in comuuter technology these
are likely to predominate in the future:^ dedicated computer
has the advantage that it is completely used in one function,
the costs are fixed, it is easily accessible and always available
to the project staff. When-the company's computer is not
suitable and bureaux charges are expensive a small computer
dedicated to this one function is an economic proposition.
At present dedicated computers exist for project management systems in two forms, but technological progress is
leading to their merger:
1 Microcomputer systems costing from 2,000-10,000.
2 Minicomputer systems costing from 60,000-140,000
Microcomputer-basedsystems The development of the 8 bit, 16
bit and 32 bit microcomputers and specialised project
management software for them has extended the use of
computer-based project systems to even the smallest firms.
The introduction of the 16 bit and 32 bit microcomputers
with 10 megabyte or larger hard disks, such as the IBM PC,
Sirius and C o r n s , has extended the number of activities
that can be handled to 4,000 or more, with some 16 bit micros
having multi-user system. A typical 16 bit system including
printer and software would cost 5,000-6,000 in the UK,
and be sufficiently powerful for the majority of firms and
projects. The number of modules that canbe linked together
has also increased, e.g. cost management module, such that
the distinction between micro- and mini-based project
management systems is rapidly disappearing.
300
ADVANCED PROECT MANAGEMENT
Minicomputer systems Firms involved in large-scale project

work can often justify the specialised minicomputer project
management systems such as Metier's Artemis system. The
Artemis system combines a sophisticated planning system,
capable of handling projects up to 32,000 activities, and a
relational data-base management system into an adaptable
and total integrated project management information system.
Originally the Artemis system was based on a dedicated
Hewlett Packard minicomputer, which now incorporates
IBM PC work stations, but it is also now available for some
main-frame computers.
The relational data-base concept enables applications to
be set up such as planning, procurement, cost management
and record management. These applications may at first be
independent and may be even maintained and updated
separately, but the relational data-base system can be used
to link and integrate them into an IPMIS.
Unlike the larger business computers, which require a
specially designea room environment and a data-processing
staff, microcomputers are designed to be used at an individual level. he^ are relatively inexpensive, can be as
portable as a typewriter,fit on a desk or table, with no special
electrical or environmental control requirements, and do
not need a data-processing professional to operate them.
More importantly, microcomputers are easy to use, with no
real data-processing knowledge required. Any person can
purchase, install and operate one. They can be set up in the
project office or construction site, to work on a turn-key
basis, in that those using them need not know how to operate
a microcomputer, but just know how to use the particular
application required.
Microcomputer-based project management systems are
effective practical tools for project managers and enable
even the smaller firm to use modern methods effectively. In
the coming years they will be adopted extensively in project
work and, hopefully,will lead to better project management,
planning and control. Whichever kind of interactive computing system is used, these modern computer-based systems
can overcome the previous difficulties experienced in planning and controlling projects.
Conclusion
On the smaller project, people are far more important than
systems. It is better to have a healthy team who are involved,
committed and who communicate with one another, than
the best system in the world. Even with poor systems, a good
team will manage effectively on the smaller project, whereas
good systems will not compensate for a poor team.
However, on the largerproject a healthy team is necessary
but not sufficient good systems for planning andcontrol are
essential to their effective management and given the
dimensionality of the problem, these systems must be
computer-based. Computer-based project systems got off to
a bad start with their emphasis on batch computers and
planning systems, but they are now much more effective.
Dedicated micro/minicomputers and modem packages on
large mainframe computers, can provide the necessary
integrated project management information systems required
for successful project management of the larger project
Even for the smaller project, microcomputer-basedproject
management systems are effective practical tools for the
project manager. Thus any size of firm and every project
manager can apply the methods and systems outlined in this
book
10 Planning and Control

of procurement
Innumerable projects are delayed due to late delivery of
materials or equipment, and yet many project plans only
show a single arrow or bar entitled 'delivery of equipment
and materials'. It cannot be emphasised strongly enough
that the management, planning and control of this phase of
a project is as important and should take as much effort as
that of design and construction no matter the size of the
project. Project management must extend to cover the firms
supplying material and equipment to the project to include
them in the global project control system. Essentially they
are no different to the other contractors, subcontractors or
departments working on the project, and are part of the
overall project organisation. Delays in deliveries can be kept
to a minimum by effective, organised, planned reporting
and expediting and unpleasant surprises can be avoided.
Thus the project manager must be aware of the factors that
go into an efficient procurement system, as follows.
The procurement personnel on a project have the responsibility for procuring all equipment and materials
required for that particular project. This procurement service
not only includes purchasing, but includes a complete
follow-up of every item to make sure that all materials and
equipment are in accordance with job specification, and are
delivered to the construction site on schedule. The procurement function can be considered to be made up of four
subfunctions, namely:
1 Project liaison.
2 Buying.
PLANNING AND CONTROL OF PROCUREMENT
303
3 Inspection.
4 Progressing and expediting.
Project liaison
This should involve the following functions:
1 To advise interested company functions as to current
commodity lead times, so that future requirements
can be identified and incorporated into the company's
overall plan, and engineering drawings and specifications are raised in sufficient time, as will enable
procurement to be achieved in a timely manner.
2 To maintain a register of potential suppliers of
equipment which may be required in the course of
the company's business, and to seek out new sources
of supply, processes and techniques, which may
enable the company to fulfil its obligations, more
profitably and quickly.
3 To assess in company with interested functions, the
suitability of potential suppliers as approved company sources of supply.
4 To advise regularly interested functions as to current
commodity~pric~s,
thus ensuring that accurate
information is used in the vrevaration of tenders.
5 To advise tendering departments as to suitable
commercial conditions to be included in purchase
enquiries, to ensure that suppliers' offers are compatible with the company's and theirclients' contractual requirements.
6 To assess commercially and adjust as necessary,
project tenders which have met the company's
engineering requirements, and to recommend which
offers should be incorporated in the overall tender.
7 To maintain such offers in a viable form, until such
time as it is established that a formal contract will
ensure and they are passed to the buyer to formalise.
8 To co-operate with engineering in the preparation of
technical specifications and requisitions for all
materials and equipment; that is, the technical

preparation of orders. When bids are received this
function also involves the preparation of bid
summaries, analysis and purchase recommendations, and is normally carried out by either or both
the engineering design groups and project staff.
Buying
The buying group is responsible for the obtaining of bids
from suppliers and the placing of orders. There are four
main types of purchase involved.
1 Equipment which is designed by a supplier, for
example, such things as pumps or heat exchangers.
2 Equipment which is either designed by the contractor
or made in the company, for example, pressure
vessel.
3 Bulk materials such as piping, concrete, sand, reinforcing steel.
4 Services such as consultants, subcontractors or
inspectors.
The purchasing department can be organised in two ways. It

can either be set up so that a group of buyers is assigned to a
particular project, or it can be set up such that specialists
concentrate on their own particular type of equipment,
material or service, and one buyer serves all the projects for
one particular type of service. The functions of the buying
group are thus
1 To determine, in conjunction with engineering,

suitable suppliers of material or equipment not
established during the project stage.
2 To incorporate into formal purchase enquiries, all
engineering quality control and commercial specifications, and to issue these enquiries to the selected
potential suppliers.
3 To receive and evaluate commercially offers made
by suppliers, which meet the company's engineering
requirements.
305
4 To check the offers received against the purchase

allowance, and to negotiate, where necessary with
the suvvlier. to ensure that the offer accevted
repres&ts the best buy obtainable by the company.
Where the best offer exceeds the vurchase allowance,
the buyer must advise interesteh functions as to the
reasons for the excess, and either obtain modification of the company's requirement, or obtain
authorisation to overspend.
5 To prepare and issue suitable purchase orders
embodying the company's requirements in respect
of delivery or completion, engineering planning,
quality control, inspection and commercial aspects
of the offers selected, either during the tendering
and negotiating, or contract stages of the project
6 To ensure that the number of informal instructions
to proceed are kept to a minimum, and that when
used, are properly authorised and formalised as
quickly as possible by the issue of an official order.
7 To off-load to suitable suppliers, work which the
company finds it is unable to complete through its
normal resources, including where appropriate the
provision of the material necessary to complete this
work.
8 To ensure that the obligations accepted by suppliers
are correctly fulfilled, including those requiring
execution of work on sites removed from their
normal places of business.
9 To resolve delays involving the manufacture and
delivery of supplier's equipment beyond the scope
of the expediting functions terms of reference,
including negotiating and authorising payment of
premium time or bonuses.
10 To determine and negotiate where necessary,
appropriate rates of pay and material costs to be
paid for additional workcarried out at the company's
request, where it is not possible to assess costs prior
to work commencing.
11 To impose on defaulting suppliers, the fiscal
commercial conditions of the order such as liquid-
ated damages for late delivery, or penalties in

respect of deficiencies in the guaranteed performance.
Inspection
This involves the physical inspection of equipment and

materials to ensure that they comply with the purchase
orders, the company's specifications and statutory regulations. The inspection department may be a separate department, part of purchasing or engineering design, but however
it is organised it must work closely with these areas and the
project staff.
Progressing and expediting
This function involves the continuous review of the performance of suppliers and relevant subcontractors. It involves
the reporting of the status of all orders, from order placement
to delivery and the exerting of pressure, as necessary, to
ensure the maintenance of planned performance. This
normally comes under the supervision of the purchasing
department, although project management often get involved.
The functions of progressing and expediting include
1 To determine whether the purchase order should be
subjected to routine progressing from within the
company, or to visits by expediting staff. This decision
will normally be reached after consideration of the
complexity, value and importance of the order and
will take into account the supplier's past delivery
record.
2 To obtain from the supplier a formal delivery
commitment.
3 To obtain where required, details of the supplier's
manufacturing plans to supplement the company's
overall plan, and to monitor the progress of the work
during manufacture.
307
4 To monitor suppliers during the life of the order, and

to eliminate or minimise, in conjunction with the
supplier, any disruptions or delays to their acknowledged commitment.
5 To determine any restraints attributable to the company or its client, and to assist the supplier in
removing same.
6 To issue delivery instructions to suppliers when
orders are complete.
7 To advise interested parties regularly, as to the status
of orders, to ensure that where necessary remedial
action is promptly taken.
The basic information which should be available for the
procurement control cycle is:
1 Delivery dates quoted on orders.

2 Schedules or manufacturer's plans for fabrication
and delivery.
3 Periodic progress reports.
Unfortunately, experience has proved that all these factors
can be very unreliable. The normal pattern of a delivery
delay is for reports on progress to be satisfactory until, as the
delivery date approaches, it becomes obvious it is not going
to be met. It is generally not possible to avoid all delays in
deliveries, but they can be kept to a minimum. Almost as
important, is the fact that, if possible, there should be a
realistic forecast of all deliveries. In many cases delivery
dates are quoted optimistically, or are based upon inadequate
estimates: many manufacturers do not produce shop
schedules, and progress reports are often obtained without
the actual work being observed.
It is difficult enough to get the project manager's own
organisation to plan and control effectively, but it is much
more difficult to get realistic plans and reports from some
manufacturers and subcontractors who are separate companies over which the project manager has no direct control.
The procurement control process must start when the list
of suppliers for bidding purposes is made up. The project
manager and other in-company groups involved should use
308
up-to-date assessments of the previous and current performance of suppliers, and select, as far as possible, those
who have a reliable performance.
Thus the procurement or purchasing department must
continually examine both existing and potential sources of
supply to ensure that they are
1 Technically competent.
2 Have adequate facilities and capacity, consistent
with carrying out the company's requirements.
3 Have sufficient inspection and quality control
personnel, to ensure that the required specification is
adhered to, and that no extensive acceptance of
corrective procedures would be required by the
company.
4 Financially sound.
To ensure that these principles are fulfilled usually entails

visiting the supplier's premises to interview key personnel
and examine the facilities offered; a process which can
range fromvisits by a commercially-oriented engineer in the
case of small firms, to visits by teams of qualified personnel
in the case of large manufacturers. This process is sometimes
known as vendor auditing or supplier evaluation and
normally involves examination of the following areas:
1 Organisation and finance.
2 Sales.
3 Engineeringkrawing office control.
4 Purchasing/stores management control.
5 Production control.
6 Production facilities.
7 Quality control and inspection facilities.
The full scale examination of suppliers usually involves
personnel from purchasing, engineering, quality control
and inspection. To ensure that such examinations are
consistent in the results they achieve, most companies
examine against a pro-forma schedule, which lists the points
on which it requires assurance and which are laid out in
such a manner, as will enable the team to complete it as the
309
examination proceeds. Such forms vary from company to

company, but normally cover such areas as:
1 General information
(a) Services offered.

( b ) Facilities available, for example, heat treatment,
research and development, non-destructive testing.
Subcontractors
utilised and the principal operations
(c)
carried out by them.
(b) Recent major customers.
(e) Labour relations.
2 Organisation and finance
(a) Organisation structure and personnel.

(b) Finance, company accounts and ability to finance
orders.
3 Sales
(a) Personnel.
(b) Points of contact.
(c) Quotation procedures.
(6) Standard conditions of sale.
(e) Willingness and ability to comply with purchase
conditions.
V) Receipt of order procedure.
\
4 Engineerinddrawing ofice procedure

(b) Engineeringldrawing control, how are client's
requirements passed to production and purchasing,
and how are revisions to these requirements handled?
5 Purchasing/stores management control
310
(b) Purchasing, how are suppliers selected and requirements made known, how are materials accepted and
what inspections are carried out?
(c) Stores management, who controls, how are stock
levels monitored and allocated?
(d) Stores control, howlaid out, how are special materials
and welding consumables stored, what quality control
provisions are taken, is any equipment used to verify
raw materials, how are stores records maintained,
what procedures are used for issue and receipt of
materials?
6 Production control
(b) Order information, what information is received by
production control about orders received?
( c ) Factory loading, how assessed and what are key
resources?
(d) Scheduling, how is delivery confirmed, are orders
scheduled on each resource, how is process recorded?
(e) Progressing, how is progress monitored and who
should be contacted, what remedial measures are
used?
7 Productionfacilities
(b) Facilities, details of bays, cranage and equipment,
heat treatment and hydraulic test.
Following an examination, the supplier should always be
advised of the team's findings and conclusions to enable
him either to correct deficiences found, or to contest the
findings reached. Although initially it is necessary to seek
permission to carry out a quality examination and to ensure
that key personnel will be available for interview, ideally
suppliers should be asked to accept the principle of random
checks to ensure that no special precautions are taken.
Audit examinations provide purchasing with the basic
information regarding a supplier's performance, but there
311
are other ways by which the company updates its information, a few of which are
1 Information gleaned from visiting salesmen regarding new products, processes, orders received and
changes in lead times.
2 Reports in technical and national press.
3 Research and technical information passed to learned
societies.
Exchange of information
Once the order is placed it is often overlooked that there is
generally the necessity of an exchange of information
between the ordering company and supplier. Equipment
drawings are often required from the supplier before some of
the design work can continue. Also further design data is
usually required by the vendor before work on the order can
actually start This exchange of information is required
quickly to avoid delays on both sides. One of the first jobs of
the expeditor may be to expedite his drawings. Similarly the
information required by the vendor often needs to be
expedited by the project staff. When criticised for late
delivery a supplier will often point out that delivery was
quoted as 'after all information has been supplied' and that
the ordering company was lackadaisical in supplying it.
Effective expediting
The function of progressing and expediting, is to achieve
planned deliveries, to give reassurance, that deliveries will
be met, and if not, to say why not and when they will be
achieved. It begins workwhen the formal order is placed and
only ceases, when the supplier has completed all deliveries.
Effective expediting requires considerable effort and must
be a positive, aggressive function. To achieve results it must
recognise the beginning of slippages in suppliers' programmes, anticipate problems and take evasive action
before delays occur. It should influence decisions before
orders are placed and start immediately afterwards, and
312
should only finish when every item has been delivered

including spare parts.
It is very desirable that the expeditors should be experienced
or technically qualified in the areas they are working in, and
have knowledge of production techniques involved. Otherwise it is not possible for them to make their own judgement
and evaluate the estimates and promises given by the suppliers.
This desirable state of affairs is very difficult to achieve,
because of the availability of this kind of personnel, and
many expeditors are simply clerks, who are not able to make
their own estimates from their observations on the shop
floor. They do their expediting by phone or letter, and when
they do visit a supplier, they find it difficult to judge his
statements. This concentration on the use of the telephone
or letters must be avoided, if expediting is to be effective. It
can be used to supplement vists to suppliers, but if used by
itself it will often lead to delays and broken delivery
promises. Suppliers, if not deliberately misleading, will
always tend to be optimistic about meeting delivery dates,
and there is no substitute for intelligent, objective observations on the shop floor. Thus the decision whether to
progress from 'in company' or to expedite by 'visit' is
reached, after considering the following factors.
I The order value.
2 The complexity of the order.
3 How critical its delivery is to the overall programme.
4 The supplier's past delivery record.
In practice, orders normally fall into the following categories
1 Those progressed by telephone or letter: (a)uncomplicated orders, (b) medium to low priority orders, and
(c) orders with reliable suppliers.
2 Those expedited by visits: (a)complex or high value
orders, (b)orders with critical deliveries or penalties
for late delivery, (c) orders subject to progress payment by result, and (d) orders on suppliers with a
poor delivery record.
One of the advantages of using both progressing and

expediting, is that the manner in which orders are moni-
313
tored, can be altered at any time up to completion to

accommodate changing circumstances such as sudden
unreliability. Progressing and expediting orders involves the
following procedures.
1 Progressing
Having taken the decision to progress, the order is passed to
the progress clerk responsible for that supplier under the
direction of a section leader whose duties are:
(a) to familiarise himselfwith the order and its requirement;
( b ) to commit the supplier if not already committed, to a
specific delivery:
to
set up lines o~communicationwith the supplier, and
(c)
obtain and vass to interested varties at the svecified
times, reports of the order's
(d) to identify restraints and discrepancies and initiate
remedial action;
to
receive sub-orders, test certificates and advice notes
(e)
and pass these to planning, engineering, inspection
and shipping;
to
advise the supplier of the address to which the
V)
material or equipment should be sent.
2 Expediting
Once the decision to expedite has been taken, the expeditor
must:
(a) understand the requirement, the supplier's structure
and organisation and his past delivery record;
( b ) make arrangements to visit the supplier to set up the
procedures to be used to monitor completion;
on
visiting the supplier, set out to generate a relationship
(c)
of mutual respect, establish points of routine and
emergency contact and decide, in conjunction with-the
supplier, how his manufacturing programme should
best be monitored;
(d) on subsequent visits the expeditor must visit the shop
floor to check progress, identify restraints and aid the
314
supplier to mount any recovery plan agreed as being

required;
(e) havingvisited the supplier, the expeditormust prompt19
submit his visit report, showing restraints and reasons
for delays. As most suppliers have a mixture of routine
and special orders in hand at any time, reports need not
always be required in the same detail.
As will be apparent from the above, the basic requirements
of both progressing and expediting are the earliest possible
identification and notification of restraints and the initiation
of remedial action. Remedial action can vary from assisting
a supplier to plan recovery, to clearing restraints within the
company, so it is essential that its 'in company' lines of
communication are as good as those established with the
supplier. Progress and expediting should n o t however,
appear internally to identify itself solely with suppliers, but
should also project the image that it is willing and able to
assist company personnel in resolving problems they are
experiencing with suppliers.
One of the problems which a progressing and expediting
organisation in heavy engineering must meet and overcome
is that while reporting by supplier. they must translate the
information gleaned into contract reports, involving many
different suppliers and differing types of materials and
equipment. To achieve this, it is becoming commonplace for
progressing and expediting organisations to include a section
responsible for the compilation and transmission of contract
reports from the information passed to them by both
progress clerks and expeditors.
Supplier's plans
Many of the problems of effective progressing and expediting
can be overcome if realistic plans can be obtained from the
supplier. Then all that is required on a visit is to compare
progress against plan. If a slippage is evident then expediting
can be carried out directly or it can be brought to the attention of the-project manager for him to apply pressure at a
more senior level. These plans, job or shop schedules as they
PLANNING AN D CONTROL O F PROCUREMENT
315
are sometimes called, should be part of the job order or

subcontract. When an order is placed it should be possible
for every supplier to produce a plan with key data on it as to
when and how he intends to produce the item ordered. Many
firms include this request in their order specifications.
Unfortunately, there is often great difficulty in achieving this
state. Planning is a n area of weakness in many firms - some
will not and others cannot - but the net result is often that
plans are not supplied, or they are extremely rudimentary, or
they bear no relation to how the job will be carried out. To
avoid delays in delivery becoming apparent only near to
delivery date, a minimum form of plan must be obtained for
all important items, materials and subcontracts so that a
control system can be set up.
It is at this stage that, if satisfactory plans are not
forthcoming, the project manager or one of his staff should
intervene. Obtaining satisfactory manufacturing plans is as
much part of the project manager's function as planning
design and construction. There are normally only between
ten and twenty critical suppliers, even on large projects, and
the problem is not as big as it first seems. The project
manager, or his deputy, should establish personal contact by
visiting the suppliers with the purchasing officer or expeditor.
On this visit it is usually possible for him to get or put in hand
the preparation of a minimum satisfactory plan. Often it is
worthwhile for the project manager to provide assistance or
guidance to the supplier in producing this plan. Two of the
planning techniques covered are very appropriate to this
stage of procurement and they both produce meaningful job
plans, which can be used for monitoring purposes.
The first of these is milestone reporting, where, as stated
before, several significant dates or milestones are established
with the supplier. These are points of significant and
observable accomplishment in his shop, such as the start or
completion of various parts of the job. They may include
such things as lead time before actual work starts; when
drawings are ready; when critical materials such as castings
are received; when various parts of the manufacturing
process are complete and so on. Thereafterprogress between
milestones, and when milestones are due to be reached, can
316
be reported by anyone visiting, either expeditor, inspector or

project engineer. In addition, telephone calls to report
progress between visits tend to give much more reliable
information when they are based upon actual milestones
being reached. The big advantage of this method is its
simplicity, especially where suppliers are unwilling to plan
or commit themselves. Any job can be split into a minimum
of four parts and delays in manufacture can be picked up on
any one part and action taken. This avoids delays not being
picked up on the overall job, and delays can only occur on a
quarter of the job before pressure can be brought to bear.
The other form of planning that can be used is to use
standard job plans, either separately or with milestone
reporting. The manufacture of most items is usually done in
a repeatable way and an arrow diagram orbar chart for each
class can be a standard sub-plan or module. Once made up,
they can be used on many suppliers for many projects, as a
basis for discussion and preparation of their plan or
establishment of milestones.
It must be emphasised that this planning, reporting and
expediting system must also be carried out with the supplier's
subcontractors and material suppliers. In many cases
suppliers give as a reason for delay the late delivery of
equipment or materials to them. Thus, to control completely
the project at procurement stage, the control system must be
extended to cover these firms.
Analysis and reports

Once the planningand control system plans are established,
reports on work progress start coming in from firms,
expeditors, inspectors and others. Progress is compared to
plan and any deviations brought to the attention of the
project manager or his staff for action. These reports are
formed into the monthly status report covering all orders; in
cases of projects with short life spans, very critical items or
items on which delivery date is slipping; these reports are
often produced bi-weekly. The report in itself serves no
purpose and is only a link in the control cycle. The same
317
elements are present in this cycle as in the project planning

and cost cycle, namely plan, report, forecast, analysis,
control action and check the effect of control action. The
analysis of the order status report and the forecasting of
possible delays are thus of extreme importance, and must
include a critical questioning of the reliability of the data
that went into the report. It must not only look for delays but,
more importantly, look for the first signs of delays and take
action to minimise or stop them.
On large projects this order status or report is often
produced by a computer, and can be several inches thick. In
these cases reporting by exception is normally used, and a
summary sheet shows the position of those items on the
critical path, and those with delays forecast. Unfortunately
experience has shown that reporting by exception must be
treated with caution, as often it is the unexpected that causes
trouble.
I n the busy life of a project manager reviews in depth are
generally not carried out each month, but just as in planning
and control of other work, there has proved to be a danger in
only following critical paths or items. Inevitably it seems
some items, thought non-critical, eventually become critical
and affect the whole project unless they are given some
management attention. Thus, in addition to a normal
monthly review, major reviews of procurement should be
carried out at logical intervals in long time scale projects, in
the same way progress and cost reviews are carried out.
These major reviews go over every order, examine the facts,
and determine if any further follow up or action is required.
The object is to catch those items off the critical path which
may cause trouble, and to ensure that the normal expediting
process is working.
In the normal monthly review, the additional technique of
sampling can be used with advantage, just as with progress
and cost analysis. Each month, in addition to the summary
sheet, a nuniber of orders are examined in depth on a
random basis to ensure that the overall system is working
correctly. This would involve not only a sample from each
type of order, but each month examining one type of order in
depth. These techniques of exception reporting, major
318
review and sampling are necessary, due to the complexity

and size of projects, and the information explosion. Unless
critical human analysis and judgement are used a computer
can simply be a big typewriter which makes unreliable
information look impressive.
If a procurement system, such as outlined here. is carried
out, it should not only have the direct effect of minimising
delays and giving accurate reporting, but it will also have
indirect advantages. Once it is seen that an interest is being
taken in procurement plans and reports, personnel involved
will give them more weight and attention. When an efficient
system is being used, suppliers will tend to co-operate; they
will tend to perform planning and reporting better than
average on the order placed, either because they see it as
worthwhile, or know that any neglect will be picked up.
11 Human Behaviour in
the Project Setting
The bulk of this book up to now has been concerned with
project management systems. There is one system which has
not been covered and without it being effective, all the other
systems will not work; that is the 'people system' or the
human relations system. Effective project systems and good
human relations systems are both necessary for good
performance on projects, but neither is sufficient on its own.
However, if one has good human relations on a project
reasonable performance can be obtained, whereas if people
are forever antagonising each other good project performance will never be achieved, no matter how good the
project systems are.
Technical problems on a project can always be solved
given time and money, but people problems are much more
difficult if not impossible to 'solve' in the short life span of a
project. Thus in addition to his professional skill in planning
and control, a project manager must develop his skills in
managing people if he is to be successful, as these are critical
to project performance, which in reality is actually people
performance. In this he can learn a lot from the work done in
the field of the behavioural sciences and also in its a articular
application to the project setting. This is because project
management is an area of management where an individual
can quickly develop and learn &man relation skills, as the
normal patterns of human behaviour in management are
both accelerated and accentuated in the management of
projects. If an individual is aware of the modem human
relations concepts, more can be learned in two years of
320
project management experience than in ten years of normal

management experience.
This is because the project manager has a very difficult
task and more problems in human relations than the
normal manager. He has problems arising out of -the
complexities and ambiguities of the matrix organisation,
such as dual subordination and uncertain authority, and
any interpersonnel, group and intergroup problems are
accentuated by the temporary nature of the project and the
fact that many organisations are involved in one endeavour.
The project manager has to manage a complex organisation requiring the work and contribution of many people
from different professional backgrounds, different trades,
different departments and different companies. Therefore
even in the best of circumstances, the management of a
project is a difficult problem. It is a very difficult task to coordinate, communicate, provide leadership, motivate all
the personnel involved, provide the necessary drive to
achieve success, to plan, organise and control, that is, to
manage a project.
A project manager's taskis made more difficult by the fact
that he acts as a junior general manager with normally illdefined authority limited to that project. His lines of
authority or influence are grafted on to the existing pyramid
structure, cutting across the normal vertical lines of command
and departmental boundaries. He does not work in the
normal superior-subordinate relationships but must manage
his peers, juniors and superiors in other departments and
companies contributing to his project. Personnel in these
departments work for two managers, namely the project
manager and their own departmental or company manager.
Thus in the normal project organisation, personnel and
groups from several different functional departments, contractors and subcontractors must work together, but are
generally not responsible to the project manager for pay
rises, promotions or performance assessments and other
line relationships. They have different loyalties and objectives, and have probably never worked together before and
may never do so again. The only thing that binds them
together is the project and the project organisation. The
HUMAN BEHAVIOUR IN THE PROJECT SETTING
321
project manager must therefore deal with the managerial

problems of developing a project team out of these diverse
groups working on a project. This involves complex
relationships with many other managers in these departments and companies.
The temporary nature of the project group means also that
members of the project work together for a limited period of
time and there is no time for interpersonnel relationships to
develop into a static state as in normal line management.
Group performance is necessary from the very beginning of
a project as mistakes made and time lost at the start can
never be recovered. In addition the composition of the total
management group on a project is constantly changing with
new members joining it and the role of some older members
diminishing in importance. It is further complicated by the
fact that project managers must work under pressure to
achieve cost, time and technical targets and their functions
include applying pressure on the people and groups involved
in the project.
Two principal characteristics of project management are
the existence of dual responsibility and split authority.
These characteristics are not new, but are more clearly
recognised in the project concept Most complex industrial
organisations recognise that the classical unity of command
concept, with authority and responsibility packaged in neat
boundaries on a traditionally organisational chart, do not
portray the actual organisational reality. Most managers in
any organisation are subject to several sources of power or
influence.
Dual subordination may not be a good thing, but as long
as instructions are non-conflicting, there is no reason why a
subordinate should not receive instructions from two people:
the project manager determining what is to be done and by
when, and the functional manager controlling how and by
whom. The project manager is primarily concerned with
time, cost and coordination, and the functional manager
primarily concerned with technical performance decisions.
When there is conflict, one superior, either the project
manager or functional departmental manager, must clearly
be recognised as the one who must be initially obeyed. Any
conflict must be resolved by these managers.
322
The problems and challenges of achieving effective human

relations in the project setting could be the subject of a
complete book or books, and all that this chapter can
achieve is to introduce the reader to some of the main factors
involved. These are:
1
2
3
4
5
The authority problem in project management.

The engineer as a manager.
Interpersonal problems in the project setting.
Group behaviour in the project setting.
Conflict and its resolution in the project setting.
The authority problem in project management

The principle of organisation from which project management varies most is that responsibility should always be
coupled to corresponding authority. The project manager is
responsible for the success of failure of his project but has
only limited authority over the personnel in his own company
and others who are contributing to his project. Yet if he is to
manage or co-ordinate effectively he must have some
authoritv over them all.
The project manager must manage across functional lines
of command in his own cornvanv.
.. and across organisational
lines of command in other com~aniesto bringclogether the
activities required to accomplishthe objectivesif his project.
In a traditional bureaucratic organisation. business is conducted up and down the vertGal hierarchy. The project
manager on the other hand is more concerned with the
lateral flow of work in horizontal and diagonal directions,
than he is with flows in the vertical direction. Problems of
motivation exist for the traditional manager but these
problems are compounded for the project manager, because
the traditional leverages of hierarchical authority are not at
he must act as a focai point for
his disposal.
making project decisions, and if he is to manage his projects
and not be simply a co-ordinator, he must be given adequate
power to aCcomplish these objectives, but often this power is
not the formal authority of superior-subordinate relationship.
ow ever,
323
Except in the divisional project concept, no project

manager whose activities cut across functional lines, or
companies can have complete authority. Authority is granted
relative to many considerations. The project manager
normallv has written contractual relationshivs with other
compar6es and these companiesgenerally waht to maintain
goodwill to ensure future business. Therefore he does have a
formal basis of authority which exists in contracts or
purchase agreements. The personnel working for these other
companies can never be fully integrated into the one project
organisation, but for all practical purposes in a healthy
situation they can work together in an overall project team.
In his own company the project manager's position is
usually one of getting the job done without the line of
authority or binding contractual arrangements for controlling
this in-house work. Thus it is often more difficult for the
project manager to exert authority over his own company
departments that it is for him with outside firms. Although a
considerable amount of a project manager's authority must
devend on sources of author* other than that formallv
&en, his position will be strengthened in his own company
bv the vublication of documentation to establish his position
and what his legal authority and responsibilities are. This
will include identifying his role in managing, organising,
planning and controlling the project, and the explicit
recognition of the project organisation form used.
Project managers in the matrix form of project management are thus faced with the difficult task of obtaining
performance from others not under their direct control. To
accomplish this they must often rely on sources of power or
influence other than formal authority. The more limited the
formal authority, the greater the pressure on a project
manager to employ other means of obtaining authority.
Authority is difficult to define and worlds like 'authority',
'influence', 'leadership' and 'power' tend to be used interchangeably and in a rather loose association. It is therefore
useful to establish an understanding of what is meant by
these words, at least in the context of this book
Authority is usually defined as the formal, legal or rightful
power to command or act. As applied to the manager,
w
324
authority is the power to command others to act or not to act,

and in the traditional theory of management authority is a
right granted from the superior to the subordinate. Authority,
at least, formal authority, is a right of a person to be listened
to and obeyed. It is delegated through the media of position
descriptions, organisational titles, standard operating procedures and related policy, and is sometimes termed position power. Such a duly appointed superior will have power
over his subordinates in matters involving pay, promotion
and performance reports.
Influence on the other hand may be assumed by an
individual without legitimacy or an organisational position,
and is often termed informal authority. An individual may
exercise influence in his environment because he has
knowledge and expertise, without the documentation of
formal authority. It can thus be based on an individual's
competence and reputation, and on his or her personality.
Leadership is a particular form of influence. It is often
associated with attempts of a superior to influence the
attitudes or behaviour of his subordinate, but it can also
apply to someone other than the formal superior, who
attempts significantly to influence the behaviour of others
with whom he is involved. The distinction between influencing and leading is that the exercise of leadership will
almost always involve an attempt to influence attitudes and
behaviour, whereas not all attempts at influence involve
leadership.
Power is a concept frequently associated with authority
and is defined as the ability to determine unilaterally the
behaviour of others regardless of the basis of that ability.
The simplest way of looking as it is to consider it as the sum
total of a person's authority and influence, that is
Power = Formal authority plus influence
As power and authority tend to be interchangeable terms
in practice, this can also be expressed for any individual as
Total authority = Formal authority plus informal authority
325
A manager's power or total authority is a combination of

his formal authority and his influence, such that subordinates, peers and superiors alike accept his judgement Thus
authority is made up not only of the formal authority
granted to an individual but also from his personal effectiveness. This can be obvious when two people who have
differing amounts of influence occupy the same position in
succession. Though this formal authority is identical, their
power or total authority can vary considerably.
The project manager accomplishes his objectives through
working with personnel who are largely professional.
Consequently his use of authority must be different from
what one would expect in the simple superior-subordinate
relationships. For professional people project leadership
must include explaining the rationale of the effort as well as
the more obvious functions of planning, organising, directing
and controlling.
Unilateral decisions, dogmatic attitudes and the resort to
the authority of a hierarchical position are inconsistent with
the project environment. The project manager's job is to
search for points of agreement, to examine the situation
critically and to think reflectively and only then to take a
decision based on the superiority of his knowledge. This
rather than his organisational position is the basis for his
authority.
The effective authority of the project manager is thus
seldom autocratic. His most meaningful authority may be
based on his ability to build alliances with the other
managers involved in the project and in resolving conflict
between the various managers involved. One asset of a good
project manager is that people from different departments
and companies who work with him on a project should
regard him in their own specialised way as an asset. The
technical, engineering or scientific people may think of him
as a buffer between them and the irresponsible demands of
other people, the finance people may think of him as a
businessman, the computer people may view him as someone
who can understand them and communicatewith operations
people.
The project manager must thus earn the respect, and gain
326
authority over elements of the project which are not under

his direct formal authority. Although the project manager
may or may not have legal authority, he gets most of hiswork
done through influence and authority other than that legally
extended. The successful project manager gets things done
through co-operation of others gained in many different
ways. This may be a combination of forces such as his status
and respect enjoyed both within and outside his organisation, his persuasive abilities, his reputation and capability in
resolving opposing views, the priority of his project within
and outside the organisation, his specialised knowledge and
his rank in the organisation Voluntary co-operation is more
effective than that forced by legal power but if informal
authority does not work then formal authority is needed.
Nevertheless in the project setting, the naked use of authority
is one of the least effective ways of obtaining commitment
and performance.
The project manager by virtue of his position and his
function has tools which he can use to establish his authority
over those involved in the project Among the most
important of these are the plan and budget Provided they
are constructed with the involvement of all those concerned,
then they commit the individuals and groups involved to
specific allocations of resources and performance. The
project manager then has the authority to hold them to this
commitment. Similarly, the use of psuedo subcontracts for
organisational cost accounts and work packages also obtains
commitment and establishes the project manager's authorits..
The project manager is also at the centre ofthe information
system and this in itself can be a means of obtaining
influence. When combined with responsibility for control
analysis and reporting, it enhances the positional power of
the project manager. In addition the ability to call meetings,
to chair them and to minute them, gives the project manager
a measure of informal and formal authority. Finally, if the
project manager is involved in the formal performance
assessment of people from other functions involved in his
project, he has in effect achieved some measure of formal
authority over them.
HUMAN BEHAVIOUR I N THE PROJECT SETTING
327
Authority, organisation and managerial philosophy

The extent of the project manager's authority in his own
company is thus the deciding factor in the form of organisation actually used. As a staff co-ordinator he has very little
authority; as a divisional project manager, he has complete
line authority over his own company personnel; and in the
matrix organisation he shares it to varying degrees with the
departmental heads.
1n the case where the project manager acts as co-ordinator,
that is, basically a staff function although given line responsibilities, the project manager must pursue performance
objectives with a great deal of skill and persuasiveness and
must constantly refer back to his line manager for authority.
The situation is not inherently incorrect provided all concerned are aware of the project manager's lack of authority
and are willing occasionally to allow delay in lack of
achievement in pursuing project goals. This is not to say that
the manager with greater authority should not use skill and
persuasiveness in his operations. However, when a crisis
arises, speed in the implementation of instructions or
corrective action is required, sometimes on the inconclusive
early evidence of advance trends. The manager with some
real authority has a much better chance of success because
he can redirect personnel and money and make decisions
with the minimum of consultation. With limited authority
the project manager who is a co-ordinator must take some
considerable time to consult with supervisory authority and
must attempt to persuade other personnel on the basis of
inconclusive information.
In the normal matrix organisation, the project manager is
working with shared or indefinite legal authoritv and thus
he depinds a great deal on his ungfficial authority. The
actual working form of matrix management depends on the
resDect and lovaltv held for the individual vroiect
- manager
by &iscollea&es. ~ e f o rthe
e matrix form of project organGation is introduced. it must be ~recededby a careful analysis
of the projects to be handled, the existing organisation both
unofficial and official, the personalities involved and the
managerial philosophies of the firm. The extent of the
A
328
project manager's authoritywillvary with each organisation

and with each project manager. Experienced departmental
managers cannot be expected to share authority with a lesser
executive who is appointed as a project manager. Therefore
project managers must have the respect of their fellow
executives and the degree of authority together with relationship to functional departments spelled out. Where there
is mutual respect and confidence in each other, the
functional managers will accept a project manager's decision
as a generalist and a manager, and he must accept their
technical recommendations. The basis for this healthy
situation is the appointment of project managers who have
their fellow executives' respect and confidence and a logical
division of authority based on a careful analysis of the
situation.
Thus one factor which determines which is the most
applicable form of project organisation for any individual
company is the managerial philosophy of that company.
Each individual organisation is as distinctive and unique as
the people in it, and each has its own personality and
character. Thus the form of project organisation which
works for one firm will not necessarily work for a seemingly
similar firm, as a company's organisation and philosophy of
management must be compatible.
Any company's management philosophy evolves over
several years and reflects the personal convictions of the
senior management of the company. Therefore any changes
in organisation which are not compatible with the deeply
held views of the top men will fail. For example, the
following points will illustrate a few aspects which affect the
success or failure of any new organisational form and which
cannot be changed overnight.
1 The prevalent managerial style is historically either
autocratic, or alternatively participative.
2 Is the company conservative or will it try out new
ideas?
3 Does it encourage delegation of decision making to
lower levels of management, or persist in having
every decision referred to senior management?
HUMAN BEHAVIOIIR IN THE PROJECT S
329
4 Is personal accountability emphasised, or must all

important decisions be taken by a committee?
These factors generally represent deeply held beliefs and
convictions of senior management and are very difficult to
change. Therefore whatever type of project organisation is
used, it must not only reflect the needs of any particular
project but also the 'givens' of the situation. If powerful
functional departments or individuals are dominant, conflict
and frustration will certainly arise if a strong project
organisation is attempted. A weak project organisation
emphasising communication, co-ordination, planning and
monitoring of progress may be the best solution for such a
situation. Where a company's philosophy is participative
and encourages delegation, a strong or fully mixed matrix
organisation will achieve better results on project work It
can lead to effective team building and the generation of a
project attitude of mind with good human relations.
The engineer as a manager

Much has been written and many studies canied out
relating to the engineer's success and failure in the field of
management. It is generally recognised that many engineers
encounter problems when they reach the stage in their
career that marks the transition from engineer to manager.
They often find difficulty in overcoming problems caused by
the lack of compatibility between the purely technical role of
the engineer and the non-technical requirements of the
managerial role. This can lead to engineers as managers
having difficulties in communicating with people, having a
lack of knowledge of other disciplines because of their
narrow technical interests and feeling conflict between
their role as an engineer and their role as a manager. These
problems can often arise out of the change of working
environment from one of precision and predictability of
physical problems to one of uncertainty and rapid change of
conditions. This presents serious problems to both the
330
engineer and the organisation in which he is employed, and

can materially affect project performance.
As the majority of managers working on a project are
engineers, the project manager must understand the problems facing engineers when they perform the role of
managers. It is useful in this contect to examine the task, role
and character of an engineer in relation to the organisation
and to compare the task, role and character of the manager
with those of the engineer. This can give the project manager
a greater understanding of the interpersonal conflicts arising
from these aspects of the essentially two different professions.
The engineer
The engineer works in an environment where physical rules

and laws are his tools. His designs are based on these laws
and consequently a high degree of physical certainty is
attached to the solution of his problems. His world is one of
black and white, right or wrong, clear decisions with clear
outcomes. He generally performs tasks of a technical nature,
calling for precision, mathematical ability, the application
of well-proven, precise theory, the mechanical approach to
problem solving and the need to seek a single, lasting
solution in an environment of high certainty and little
change. The engineer's role thus requires an attitude of
precision, a mechanistic approach to problem solving and a
bias towards technical factors.
Thus the general conception of an engineer is that he is
unemotional, prefers to deal with things rather than people,
has narrow technical interests and chooses to specialise in
physical sciences. As a result, studies of the personality of the
engineer identify some of the engineer's typical personality
traits which are symptomatic of the problems encountered
by an engineer in his transition to management. Traits such
as difficulty in dealing with people, procrastination in
decision making, fear of being proved wrong, a resistance to
change, maintenance of the self-image as an engineer and
not a manager, sensitivity to criticism, a bias towards shortterm rather than long-term planning and a distinctly taskoriented behaviour.
33 1
The manager
The manager on the other hand performs the task of

planning, organising, directing and controlling the resources
of his firm, principally men, money and materials, in a world
of uncertainty, intangibles and constant change. These tasks
call for a much broader, creative approach, and ability to
diverge and search for many alternative solutions to a
problem and to then choose the best solution applicable at
that particular time. The manager's role in contrast to that of
the engineer, thus requires an attitude of precision only in so
far as its usefulness is justified for the goal to be achieved
and the time available to reach that goal. It requires a nonmechanistic approach to problem solving and a bias towards
efficient resource utilisation and more people orientation.
T%e engineer's career path
If an engineer's career path is traced through his organisation, he will start his career usually, at a point part-way into
the task-oriented portion of the engineering function. As his
career develops he will become less technically-oriented as
he is given more and more responsibility for staff and other
resources. He will become progressively more managerially
oriented in his role until, he crosses the boundary into
functional management At this stage in his career he is
concerned with managing engineers in the function or
discipline in which he has worked for a number of years, but
he needs to recognise that he is now predominantly a
manager, and not an engineer. Not all engineers who have
made this transition recognise this factor.
In the traditional non-project organisation, the next stage
of promotion is for a functional manager to move into a
general management position, where he is responsible for
the management of resources for the complete organisation.
At this stage he is concerned with multi-disciplinary
management, and thus has moved almost completely away
from his technical role. This is a very demanding task for
someone who started out as a technical specialist, but a
332
AD VANCED PROJECT MANAGEMENT
limited number of people of any discipline reach the top

position of general management.
However, in project management an engineer can reach a
position of general management at essentially the level of
function manager, that is, middle manageror lower and thus
is faced with the problems of multi-disciplinary management
at a much earlier stage in his career development.
The engineer as a functional manager
When a n engineer first becomes a functional manager he
runs up against two problems namely, role identity and
conflict. The engineer although in management, tends still
to identify himself more closely with the role of a n engineer,
rather than that of a manager.
The above attitude can lead to internal stress and can
impair the performance of the functional manager, if he
continues to bias his role backwards to the task-oriented
engineering role, leaving less time to perform his prescribed
role as a manager. This can lead to interpersonal conflict
with both project management and the engineers under the
functional manager, with both parties recognising that he is
not performing his role as a functional manager.
Similar symptoms of intergroup conflict occur between
the engineer as a functional manager and other departments
or segments of the organisation. The boundaries between
the engineering segment and all other segments are essentially potential points of conflict and give rise to such
problems as territorial jealousy, role conflict (due often to
role overlap), withholding or distorting information and
poor interdepartmental communication.
The engineer as a general or project manager
When an engineer moves into the general or project management area he is faced with several problems. He has to lose
the greater part of his identity as an engineer and has to
acquire an understanding of and sympathy with the needs of
the total role comprising all parts and functions of the
organisation. He has to obtain, or already possess the skills
HUMAN BEHAVIOUR IN THE PROJECT SE'ITING
333
needed for this new role. It is here that the engineer may find
his problems are at their greatest, due possibly to the lack of
management education and training and the incompatibility
of his versonalitv with the reauirement of his role.
~~mbtom
su'hs as differenies of perception and orientation of the different de~artmentsand oreanisations contributing to the project, tike needed for thecfask,interpersonal
orientation and environmental needs arise. The general or
project manager must master these skills and understand
these needs in order to carry out effectively the task of
managing and resolving the inter-group conflicts arising.
Engineers in general thus tend to be tremendously conscientious and work oriented. Their loyalty or commitment
is very often to the plant equipment or project they are
working on rather than to the overall company. They have a
great deal of pride in their work, but this can often be a
disadvantage to the project manager. This pride leads to the
engineer being very sensitive to criticism and having an
enormous need to be right. He is more concerned with being
technically correct than with time and cost objectives, or
with people. Promotion to managerial position often
depends on the engineer's technical abilitv. rather than the
ability to manage people, and thus the eng&eering manager
often takes with him this emuhasis on technical ~erfection.
pride in his work or group; and sensitivity to chticism.
When this is combined with the fact that in many
organisations the level of interpersonal trust, support and
co-operation is low, it means that the project manager has a
very difficult task in managing the many people and
organisations involved in his project.
Interpersonal behaviour in the project setting
The project manager has the task of overcoming the many
problems involved in obtaining his personal and group
performance from individuals and organisations in achieving
the project's objectives, within the'givens' of the managerial,
organisational and technical systems they work in, that is,
the sociotechnical system. Any individual's performance on
334
his job is a function of this sociotechnical system, his or her

ability and motivation. Given that many of these factors are
not under the project manager's control, the main behavioural
factor he can influence is the motivation of the personnel
working on his project. The skills of the project manager in
dealing with people can make a significant difference to
performance on his project, and in this an understanding of
the work of the many people who have studied the problems
of human behaviour in management is well worthwhile.
Very few organisations are ideal and a person's manage
ment history can lead to attitudes, which can perhaps in the
short term of a typical project life be modified, but rarely can
be changed. Man is a complex being highly variable, but
capable of changing. The same manager may perform
poorly in one part of the organisation, where he feels
alienated, but perform satisfactorily in another part.
Managerial styles may be required to be different for
different individuals and different groups. Human beings
are complex and no one set of assumptions is generally
applicable; different people and different groups of people
will show behaviour that can be accounted for satisfactorily
by several different sets of assumptions about people. Thus
each manager in each situation must avoid making assumptions too soon and must draw on a range of models of
human behaviour to see which one suits the behaviour he
observes around him at work.
The classic assumptions about human behaviour have
been crystallised by McGregor in his Theory X and Theory
Y ,and both may be appropriate at the same time in any one
project organisation. Theory X postulates that man is a
rational economic being with the following characteristics:
1 The average human being has an inherent dislike of
work and will avoid it if he can.

2 Because of this characteristic, most people must be
coerced, controlled, directed and threatened with
punishment to get them to put forth adequate effort
towards the achievement of organisational objectives.
3 The average human being prefers to be directed,
wishes to avoid responsibility, has relatively little
ambition and wants security above all.
HUMAN BEHAVtOUR 1N.THE PROJECT SElTING
335
Theory Y, on the other hand, postulates that

1 The expenditure of physical and mental effort in
work is as natural as play and rest.
2 External control and the threat of punishment are
not the only means of bringing about effort towards
organisational objectives. A man will exercise selfdirection and self-control in the service of objectives
to which he is committed.
3 Commitment to an objective is a function of the
rewards associated with their achievement.
4 The average human being learns, under proper
conditions, not only to accept but to seek responsibility.
5 The capacity to exercise a relatively high degree of
imagination, ingenuity and creativity in solving
organisational problems is widely, not narrowly
distributed in the population.
6 Under the conditions of modern industrial life, the
intellectual potentialitiesof the average human being
are only partly utilised.
In the project setting, the project manager does not normally
manage workers at shop floor level but is involvedwith other
managers and professional people and this concept of a selfactualising man outlined in Theory Y is probably more
realistic, given a healthy sociotechnical system. The problems
of motivating professional people has been studied by
Herzberg and others who found that accomplishment and a
feeling of job growth are genuine motivators for people like
accountants and engineers. He found that what they call
hygiene factors such as pay working conditions and canteens
could cause dissatisfaction, but their improvement did not
lead to satisfaction, merely to the elimination of dissatisfaction. They were essentially prerequisites for motivation but
did not motivate by themselves, just making it possible for
motivation to operate. Motivating factors are more associated
with the job itself. Thus self-actualising man had a range of
motives ordered in a hierarchy of importance depending on
how they contributed to his survival. Given that his basic
body and safety needs, and his desire for fellowship are met,
336
his desire for autonomy in the work he does and fulfilment

of himself in his work are such that a suitably designed job
will enable him to engage himself fully in it and so satisfy
himself and the needs of the organisation. In the project
environment where hygiene factors are generally reasonably
well cared for, the personnel can be assumed primarily
concerned with a search for fulfilment of their self-esteem
needs. In this the characteristics of an engineer may be both
a disadvantage and an advantage to the project manager.
The project manager can in this environment benefit from
the use of participative methods of management In fact in
the matrix set-up he is always forced to use participative
methods to varying extents with other functional managers
and other company's managers. Participation means a
willingness of the project manager to consult with his fellow
managers and other personnel working on the project, to
acquaint them with the project's problems and to involve
them in the decision making. The participative project
manager does not abdicate his organisational responsibilities. He is still responsible for his project, but he has
learned to delegate and share operating responsibility with
those who actually perform the work Participation increases
this opportunity for those working on the project to develop
increased ability, satisfaction and thus motivation from the
project.It can lead to the removal of conditions which set the
personnel and groups apart.
However, each individual and group involved in a project
may differ widely in their human relation problems, and in
how much they may be trusted to achieve sub-goals on their
own, or how much they require to be closely monitored. The
project manager has to analyse quickly the individuals and
groups working upon his project, and ascertain specifically
what are the best methods of dealing with them and what
motivating factors are most important to them.
He must be a good human observer and be looking for
signs and symptoms of what makes them tick right through
the life of the project The abilities and motivations of the
people in the various sub-organisations are so variable, that
he must have the sensitivity and diagnostic ability to be able
to appreciate any differences quickly.
HUMAN BEHAVlOUR IN THE PROJECT SETTING
337
It mav be necessarv to be detailed and autocratic with one

individial, delegate\i;ith mild control with another group, or
individual and be fullv ~articiuativewith others. Each
management style or strat& wilfbe the correct one for each
individual or group. The project manager should become
skilful in several different kinds of managerial styles, each
consistent with the particular situation. The authoritarian
style is more consistent with many traditional organisational
forms, whereas participative management can workwell in a
matrix organisation. To do this the project manager must be
aware of his actual managerial style and consciously switch
styles when required.
The temporary nature of the project group means the
members of the project group work together for a limited
period and there is no time for interpersonal relationships to
develop into a static state, as in normal line management.
Group performance is also necessary from the very beginning
of a project, as mistakes made and time lost at the start can
never be recovered. In addition, the composition of the
group is constantly changing with new members joining it
and the role of some older members diminishing in
importance.
He must be able to size up people quickly, and be aware of
the electricity in the air depicting tension between groups. It
is a fact that in the temporary nature of the project setting,
attitudes and relationships are'set'very quickly. The position
is made more complex as project managers must workunder
pressure to achieve cost and time objectives, and their
function includes keeping the pressure on the personnel
involved in the project. They have to gain a feel for how
much pressure can be usefully applied. Most personnel
work better under a certain amount of pressure, and indeed
accept and welcome it in the project setting in order to
achieve reasonable objectives. This comes from accepting
the new challenge and taking risks and increasing their
abilities, but too much pressure leads to resistance, conflict
and unproductive tension.
Therefore, the project manager must be sensitive to the
reactions of people and endeavour to act supportively
instead of threateningly. He must be perceptive to the
338
reaction of people to him, and the interactions between

people, and this involves such factors as being aware ofbody
language, and of the factors that indicate threatening
behaviour or supportive behaviour. For example, the first
word chosen, tones of voice, points omitted, can tend to
increase defensive reaction, restrict communication and
impair the person's commitment and motivation to the
project. Supportive behaviour from the project manager or
other people can tend to reduce defensive behaviour and
lead to greater project commitment. The project manager
does have one important factor going for him in obtaining
good personal performance. Project work by its very nature
can be very stimulating, satisfying and give to those involved
a real sense of achievement. Many professional and supervisory people feel alienated by the nature of their work and
their failure to see how it fits in the overall company picture.
The many layers of management in a large organisation
leave those at the lower levels feeling a sense of powerlessness
and remoteness from decision making, and it is difficult for
them to equate their own personal needs with that of the
organisation. This leads to a loss of involvement and
commitment to their work and the objective of the groups
they belong to.
Project work with its definite visible goals can lead to a
high level of personal commitment and satisfaction. Everyone on a project can become associated with its success or
failure. For example, the objectives of a project group may
be to design and build the physical plant required by the
company to fulfil certain market demands. With this objective in mind, the personnel involved can readily visualise
and adopt the company's objectives for their own. Projects
can be extremely frustrating for a project manager caught in
a power complex with little authority, but more often, people
in all kinds of occupations involved in a project comment on
the pleasure they had working on it. They can see how their
contribution fits in to the overall picture, and if they work a
little harder, they can see what effect it has on the progress of
a job. This kind of feedback leads to a greater interest and
motivation in the job. It leads to the development of the
project attitude of mind, in which peoples' interests are
339
subordinated to the overall project and they associate

themselves with the project.
In addition the concept of cost accounts and work
packages as pseudo subcontracts may be partially threatening in a control sense, but in fact give people clear objectives
and clearly delegated responsibility.This gives them personal
targets and essentially is equivalent to the methods used in
management by objectives, and can lead to a greater
commitment to the overall project and their part of it, which
is clearly defined.
The project manager must thus make use of this unique
motivational opportunity to obtain personal, group and
global organisation performance. He must deal with people,
with the assumption that most individuals have drives
towards personal growth and development, and these are
most likely to be actualised in an environmentwhich is both
supportive and challenging. Most people desire to make and
are capable of making, a much higher level of contribution
to the attainment of organisation goals than most organisational environments will permit. However, he is working
with many different people from many different organisations, and therefore with the 'givens' of any situation he may
have to treat different individuals and different groups in
entirely different ways to achieve the best performance for
his project.
Groups in the project setting
The great majority of human beings have a natural desire to
seek the companionship of other human beings in informal
or social groups. They seek this for basic personal and
psychological reasons such as satisfying human needs for
friendship, for association with others, for a sense of
increased security and as a home base. Thus most people
wish to be accepted and to interact co-operatively with at
least one small reference group, and usually with more than
one group. One of the most psychologically relevant
reference groups for most people is the work group which
includes peers, superiors and subordinates.
340
These groups can make a considerable contribution to

achieving an organisation's objectives in that they can give a
member of a group a feeling of security and of belonging,
can maximise the contribution of each individual involved,
lead to a greater participation in decision making and
problem solving, and result in a greater commitment of the
individual to the group's and hopefully the organisation's
objectives. On the other hand, if the group is unhealthy, it
can lead to internal bickering within groups, the group's
objectives not being the organisation's objectives and to
serious conflict between groups, all ofwhich are detrimental
to achieving the organisation's and the project's objectives.
The project manager must therefore be aware of the
characteristics of groups and of intergroup behaviour and of
the factors which can lead to the evolution of effective
groups.
Principal factors contributing to the formation of a group
are:
1 They are engaged together in a task or operation.
2 The come into day-to-day contact with each other.
3 They are interdependent.
4 They have the same backgrounds, skills and sense of
values.
5 Leadership and management.
Several different types of groups have been classified and all
exist in the project setting, namely:
1 Vertical groups.
2 Horizontal groups.
3 Mixed groups.
Avertical workgroup consists of people from different levels
in the same company, department or function. Formal,
permanent, vertical work groups form part of the organisation
of most companies, and provide a home base for the
majority of people involved in a project They are normally
created to carry out or perform one specific function and are
made up of people with the same background, skills and
values, and can be considered as a uniform group.
A horizontal group in a project is a group of managers or
341
technologists at the same hierarchical level or status, often,

but not necessarily so, in the same profession. For example,
the project managers of the owner, the principal contractor
and main subcontractors may form a group by themselves,
or may form a group with the heads of the various functional
groups contributing to a project, all more or less at the same
hierarchical level. The mixed group is one which includes
people from different levels from different companies or
departments involved in the project.
Groups are very necessary in the project setting because a
project is never a one-man or one-manager task The
complexity of a project, the contributions required from the
many different specialised skills which no one man can
have, the amount of work involved, the large number of
activities that require to be managed, organised, planned
and controlled, the mass of information necessary and
normally the number of companies involved, all necessitate
the involvement of a number of managers and specialised
technologists from different functions. Thus in a large
project there is not one group but many formal and informal
groups. Likert in his bookNew Patterns ofManagement states
that organisations can be considered as systems of interlocking groups. The typical project organisation is thus not
just a set of relationships between individuals, but between
sets of interlocking and interdependent groups. These interlocking groups are connected by individuals who occupy
key positions of dual membership of groups and serve as
links between these groups. In the typical project organisation
there are many groups of all three types: each individual
department or section will tend to form a vertical group, the
people in each company working on the project will tend to
form a mixed group, and ideally, all the people working on
the project will form another mixed group. These groups
may lead to greater effectiveness, but they can also lead to a
consolidation of a 'we/they3attitude.
The vertical group, consisting of people from different
levels in the same company or department, is probably the
group that forms most naturally in the project setting, but it
may be the least desirable from the project manager's point
of view. It can be a healthy group, with great cohesion and
342
team spirit, and can lead to greater effectiveness within the

company or department on internal matters. The great
danger is that aLwe/they'attitude can be established between
this group and other functions, departments and companies
working on the project. That can lead to complex intergroup
resentment and conflict, and as all groups in a project must
work together, this in turn leads to a less effective overall
project group.
The senior members of the departments, functions and
companies in the project are usually involved on a day-today basis and form a horizontal group which can link the
overall organisation of the project together. They occupy key
roles in the project organisation and serve as channels of
communication, influence and policy integration from one
group to another. The personnel in the various groups may
not all be aware of each other, but if the horizontal linking
group is effective this will set the whole tone of the organisation and tend to lead to the other groups becoming effective
teams. It is possible for a project group involving several suborganisations with different management philosophies to
work effectively together provided that the linking members
of this horizontal subgroup form an effective team. They can
be bound together by mutual respect and an over-riding
commitment to the project.
Horizontal groups may tend to form in each function, for
example, the construction staff of the contractor will come
into contact on a day-to-day basis with the client company
construction staff. This may lead to conflict if each forms a
separate group with the 'we/they' attitude, but if they can
combine to form one group, this will lead to greater overall
effectiveness. This does not mean that there will not be
conflict, but can mean that conflict can be resolved by the
logic of the situation. It involves mutual respect of both ability
and commitment to their common objectives. Among the
factors affecting the chance of achieving this type of group is
the simple one of physical contact Take the construction
group, if each has a separate site office the tendency is to
form separate groups. If they share the same building, have
day-to-day contact, have lunch and coffee together, there is a
much greaterlikelihood oftheir forming one effectivegroup.
343
The principal group the project manager must endeavour

to establish is a mixed group, which includes people from
different levels and different functions of the departments
and companies involved in the project. This kind of group
leads to greater cohesion of the total project group and a
commitment to the total project objectives. However, this is
probably the most difficult group to create as the project
group is a temporary formal group created to carry out a
specific project, and when the project is finished, the group is
broken up. The group may exist for several years, but the fact
that it is temporary has an influence on each member. They
may feel the group may be broken up at any time, and are
more or less looking over their shoulders all the time to their
parent company or department
The members of this formal project group carry out the
formal tasks necessary for the completion of the project and
this can be regarded as the basic function of the group. If this
is the only function of the project group then it will not
develop into an effective team. Hand in hand with the
formal functions must go the informal functions of satisfying
the behavioural, personal or psychological needs of the
group members. The objective of the project manager is to
get the aims of the informal groups and formal groups in the
project committed to the same objectives. If he fails on this,
the informal groups may be hostile or apathetic to the formal
project goals and waste time and energy on intergroup
conflict If he succeeds, the project is on the road to success if
it is humanly possible.
There is a great difference between a number of individuals
working together and an effective group or managerial team.
One needs only to note some of the observable signs of an
effective team namely, team spirit, enthusiasm for the
project the members are supportive of one another and use
the term 'we' instead of '1'; they manifest towards one
another, if not friendliness at least respect for other members'
competence and points of view: they show staying qualities
when things get rough and a resistance to frustration; there is
a minimum of bickering and members do things because
they want to, rather than because they have to.An ineffective
group will have apathy, jealousy, bickering, disjointed effort
344
and pessimism about the project Instead of saying 'we will

have a go and make it1, members will point out all the
difficulties involved and will be negative about achievement.
If the co-operative relationships and interdependence of
teamwork can be achieved, the interest and enthusiasm of
all concerned, regardless of their organisational responsibilities will be committed to the project. The great advantage
of an effective group in project organisation is that greater
emphasis is given to the total organisational effectiveness
than to that of the departmental or individual companies
contributing to the project. Objectives between companies
will differ, but all will have a commitment to the success of
the project.
Team development
Effective m
s teams do not just coalesce as soon as
- o u ~ or
people are brought together. ~ r o u are
~ sdynamic entities
which are not static and unchanging. Studies of groups have
shown four stages in the processof formation ofan effective
group. Effective in this instance means working towards
project objectives, without interpersonal or intergroup
conflict, and with satisfied and involved members. These
stages are:
*
1 The development of mutual acceptance and trust

leading to a diminishing of defensive behaviour.
2 Open communication.
3 Co-operation and sustained productivity.
4 Resolving of problems and control by mutual
agreement
Thus an effective group takes time to develop and at any
stage in process can go in reverse. That is, an effective team
can be destroyed by many factors; unsupportive management philosophy of any contributing company, a rogue key
member, or failure of the project Once a team has reached
full effectiveness, however, it is reasonably resilient to minor
factors affecting it. Such a team is a good training group for
new staff and has also been used to endeavour to correct the
345
behavioural practices of problem members of staff, provided

that they are not key linking members.
Thus for a project to be successful, the project manager
must provide the leadership necessary to develop a managerial team and the organisational environment must be
favourable to his work Where a company's philosophy is
participative, a strong or fully mixed matrix organisation
will achieve good results. It can lead to effective team
building and generation of a 'project attitude' with wide
behavioural consequences. This project attitude is a way of
thinking that penetrates throughout the organisation and
unifies all activities towards accomplishment ofthe project's
common goals. It will become no longer enough to say 'that
our department's effort was satisfactory but that the project
was held up because of someone else'. No single organisation
effort is satisfactory in a project unless the project is a
success, and every effort should be made to assist other
organisations to cany out their tasks successfully. It involves
taking off departmental 'blinkers' and co-operating by
helping one another to complete the project successfully in
terms of all its objectives.
In order to build an effective team and generate this
project attitude, the company's managerial philosophy must
permit the project manager touse participative management
with its emphasis on creativity, open communication and
participation. The project manager has also thus to be more
aware of human relations and personal sensitivity than the
normal functional manager.
The leadership capabilities of the project manager are
essential to the growth of an effective group, that is, the
personal qualifications, skill role and strategy of the project
manager. The development of mutual acceptance, trust, cooperation and open communication can be helped or
hindered by the project manager, but cannot be brought
about by him alone. Nevertheless an essential factor leading
to an effective group is the perceptiveness and the ability of
the project manager to recognise and resolve group problems. The sort of signs the group manager must be on the
lookout for are, for example, members of a group at a project
meeting who fail to pay attention to each other and show
346
lack of respect, members preoccupied with their own personal objectives, the existence of threatening attitudes, etc.
For a group to optimise its effectiveness, the project
manager cannot perform all of the leadership functions in
all circumstances at all times, and all group members must
assist each other with effective leadership and member
behaviour. Thus the behaviour of the group
. is not only
influenced by the project manager's leadership qualities, but
by attitude, knowledge, skills and capabilities of the individual
members and the characteristici of the companies and
departments involved in the project.
Thus the project manager must deal with the managerial
problems of developing a project team out of these different
groups working on a project. In this he is helped by the fact
that the main advantage of project-type workis that there are
clearly definable goals of schedules, cost and performance,
which are ideal for managerial team building. If the cooperative relationships and interdependence of teamwork
can be achieved then the associated intrinsic rewards
associated with a successful project can generate the interest
and enthusiasm of all concerned, regardless of their organisational responsibilities. In achieving a project attitude of
mind the project manager has one distinct5dvantage in that
the project is a discrete entity in itself. Thus the group has
one primary task, which is the completion of a clearly
defined project, within a specified time span, to a technical
specification and within a budgeted cost. People working on
the project are associated with something concrete, not
something intangible, and a successfully completed project
is obvious to everyone and people even remotely connected
with it can draw satisfaction from this. The project manager
has to use these advantages to establish effective team work
and all that arises from it.
The project also has the advantage that all contributors to
it must work together in establishing plans and budget and
in carrying out the work The project manager cannot
arbitrarily set time limits and budgets; he must negotiate
what is reasonable with the various functional managers
and companies involved, making trade-offs between time
and money in the light of the overall situation which only he
347
knows, but which he must communicate. There is also the

opportunity for intrinsic rewards which are so important to
professional people and which exist in the clear and
unclouded achievement of the project The complaint of
members of functional groups is often that they never see the
overall picture but only their small task. It is possible in
project work to communicate the whole picture and for the
individual members to share in the success or perhaps the
failure of the project. Extrinsic salary increases and promotions can then be fairly based on project contribution and
achievement with the assessment of personnel normally
being carried out both by the functional department head
and the project manager. Co-operative relationships and
interdependence thus must exist between all those working
on a project or the organisation would grind to a halt. This in
itself can lead to effective teamwork
Finally, the more successful the management of a project
is, the more obvious is the attitude of mind of those involved.
They have a very real confidence in themselves and in each
other. This creates an atmosphere of mutual respect and
collaboration, which is essential when a large group of
people must work together towards a common objective.
Thus, though the project manager is faced with the
problem of welding together people from various backgrounds and organisations into one, effective mixed project
team, he does have certain advantages which can help him
overcQme the disadvantages of the matrix organisation. It is
up to him to make use of these to achieve the project's
objectives. Unfortunately because of the existence of
different objectives and the fact that many groups are
involved in a typical project, he must also have great skill in
overcoming or preventing intergroup conflict
Intergroup conflict in the project setting

It is an unfortunate fact that in many medium to large firms,
even with the normal functional and hierarchical organisation, there tends to be intergroup conflict or hostility.-1t is
almost impossible to have a project without differences
348
between people; differences of opinion, values, objectives,

etc. These differences can lead to discussion, argument,
competition and conflict. Discussion and argument are
constructive, whereas competition can be both constructive
and destructive, but conflict is always destructive. It would
be foolish to imagine that groups can work together on a
project without disagreement and some conflict. Disagreement between people is almost inevitable in the project
setting, and indeed, to a certain extent, it would be undesirable if this were not so. There is bound to be some
disagreement in a healthy organisation, and it is often
essential for efficiency and effective decision making.
One of the most important functions of the project
manager is thus the solution of intergroup conflicts occuning
among groups involved in his project If he is to maintain an
effective mixed team comprising all the groups working on
his project, he must constructively prevent and overcome
this destructive intergroup conflict. If he does not, the project
performance will almost certainly be seriously affected.
Conflict in a project arising between
I Individuals in the same group.
2 Individuals in different groups and companies.
3 Groups in the same company.
4 Groups in different companies.
5 Between companies, that is, between sets of groups.
In particular there is the great likelihood of conflict
between
1 Project and functional groups.
2 Engineering and operations groups.
3 Owner and contractor groups.
Consequences of conflict
Where there is hostility or conflict between key linking
members of groups, or between groups, it breaks up, or
makes impossible to form a healthy effective team out of all
those working on a project. It leads to a lack of respect and
trust between groups, a lack of harmony and co-operation,
349
and a breakdown in communication, with information

being distorted, censored or held back Each group will tend
to reject ideas, opinions and suggestions arising from the
other groups, and feelings or emotion will run high with a
greater chance of mistakes being made by people under
stress with clouded judgement Groups will tend to have
unspoken objectives, different from those of the project,
such as to'get' the other group, block anything they propose,
achieve dominance over them and show them in a poor light
to senior management Project objectives will be subordinated to the group goals which concentrate on achieving
dominance or victory over other groups. This accelerates the
breakdown in communication between groups, and creates
unfavourable attitudes and images of other groups.
There will be a polarisation into a 'we/they' attitude,
instead of 'all for one and one for all', Decision making and
problem solving will be slow and difficult, differences will
not be worked through in an open manner and youwill have
widlose situations leading to more conflict and hostility,
lowest common denominator compromises, or submission
of disputes to higher levels of management for arbitration.
In general conflict is detrimental to overall project performance and will make it almost impossible for a project
attitude to develop.
On the other hand, conflict between groups can actually
enhance the cohesion and team spirit of the individual
group. Competition does stimulate individuals and groups
to greater performance in a similar way that it stimulates an
athlete to give of his best. Group loyalty will increase,
differences will be buried within the group and there will be
a greater commitment to the group's objective, but not
necessarily to the project objectives. They will tend to close
ranks against a common enemy, that is the other groups.
Within these groups there is thus a more purposeful atmosphere, and probably more autocratic leadership patterns,
more structuring and organisation, more in-group loyalty
and conformity for a solid front to the 'enemy'. The group
within itself tends to be more effective in achieving its own
objectives where co-ordination and interaction with other
groups is not required Though this may be advantageous
350
for the individual group, it will prevent the formation of

horizontal and mixed groups and lead to poorer overall
project performance.
Reasons for conjlict
It is probable that whenever individuals are involved together

in any undertaking or operation that, given the variations
that exist in human nature, there will always be differences
and thus a potential for conflict. One of the basic reasons for
this in a company is that there is a division of labour, that is,
functionalisation, and this inherently creates groups with
different sets of values and objectives. In almost every
medium-sized and large company there tends to be to a
greater or lesser extent, antipathy or conflict between line
and staff, and between different functions, for example,
marketing and production, operations and engineering.
Each group will have a different set of values and given the
same inTormition, even with goodwill between groups, they
will often come up with different points of view, or decisions
based on these values. For example, the design engineering
groups will want to design the project to the highest
technical standards, whereas the project group will want to
compromise between time, cost and technical standards.
This will influence their points of view on many decisions
and problems and will inevitably lead to disagreement with
a potential for conflict. It will also lead to differences
between objectives between groups in the same company.
For example, the operations people will want a production
plant which is easy to operate, has many installed spare
pieces of equipment, capacity for expansion and possibly
some items which would be classified as luxuries by the
project staff. The project group will want to balance time to
completion and overall cost against performance standards,
and thus they will have different points of view, values and
objectives, and inevitably there will be differences and thus a
potential for conflict. Request for changes to design, as
discussed previously, are also a source of conflict between
the design, project, construction, manufacturing and operations groups within a company.
HUMAN BEHAVIOUR IN THE PROJECT S E T ~ N G
351
There will also tend to be competition between these

divisions of labour, that is, groups, in the same company for
dominance within the company (for example, production is
the prime department and 'what they say goes') and for
resources of men, money and promotion. There is also the
problem of priorities and competition between projects for
scarce resources. Certain groups involved in a project may
be handling senera1 projects at the same time, and their
managers may have to allocate their resources to the various
projects in a way that the project managers of the individual
projects may not like. This leads to differences, and hence
conflict between the project manager and the managers of
these functional groups, and also between project managers
in the company for these scarce resources.
This difference invalues and objectives will also extend to
the owner and contractor company groups working on a
project. The contractor is in business to make a profit and he
can only do this at the expense of the owner, or so the owner
thinks. The owner wants to minimise the cost of the contract,
which he can partially do at the expense of a contractor, or so
the contractorthinks. The extreme expression ofthis is in the
cost plus contract. In the worst situation the contractor, or so
the owner's staff consciously or subconsciously believe,
wants to maximise the cost of the project, at the expense of
the owner, to maximise his profit. Thus the owner's staff will
tend to supervise the contractor closely, question his decisions
and performance, and from the contractor's point of view,
interfere far too much. In a fixed price contract, the extra cost
of changes will be a source of friction as discussed previously.
The problems arising from the division of labour in the
normal firm are increased when the matrix organisation is
used. There tend to be many groups involved in the typical
project and there is thus the potential for conflict at each
group interface. The project management group needs to be
involved in the planning and control of the work of all
groups in the project and this can be resented. Where
responsibility and authority are unclear, there is always the
potential for conflict. Defence of territory is one of the prime
sources of conflict, and functional managers and contracto8s
managers will always tend to believe that the project
352
manager is impinging on their territory or authority, but he

must do to do his job effectively. The dual subordination of a
functional group member to the project manager and
functional manager can lead him to feel insecure in his
position, and the functional manager may also resent the
project manager's interference or, as he believes, reduction
of his authority. With a contractor, contractual relationships
may be unclear and this will always tend to-cause conflict
In a project, people are always working under pressure,
and the project manager must always be a hustler, and thus
will be applying more pressure to people to meet time and
cost objectives, and this in itself can lead to conflict. When
added to the role and territory uncertainties, it will enhance
the personal stress that people are working under, emotions
will be raised, tempers will be short and conflict can easily
arise.
When groups or companies have different managerial
philosophies, there will inevitably be differences. There will
also be personality clashes in any organisation. There also
tend to be problems with defensive behaviour by managers
whereby they are reluctant to implement change, act on their
own, delegate, take risks and make decisions. This can arise
when a manager is promoted beyond his abilities, where a
technologist has been promoted to a managerial position
and either has not adjusted to the managerial requirements,
or is now out of touch with the technology he came from and
he feels insecure. There is also occasionally a rogue manager
involved who is determined to succeed at the expense of
everyone else. He may be autocratic, ruthless and willing to
stab other managers in the back to get on. Not only does this
in itself cause conflict, but the other managers involved will
probably react to defend themselves, and teamwork will be
lost
There are also problems which by themselves in a healthy
team would not create conflict, but when coupled with one
or more of the other problems described will cause conflict.
These include differences in technical opinion, differences
in time, cost and technical standards and strong personal
commitment to one line of action.
The methods used to resolve conflict can often in them-
353
selves cause resentment and thus lead to further conflict. In

resolving differences there are several strategies a manager
or group can follow. For example:
1 Forcing through their point of view by the use of
formal or informal authority. The naked use of overriding power may gain the point in question, but it
will probably ensure a lack of commitment to carry it
out and increased feelings of resentment and hostility.
2 Submission of the differences to a high authority,
which will lead to the same result for the loser.
3 Withdrawal from the confrontation and sulk, with
the same result.
4 Compromise at a low level of agreement with the
probable result that both groups will lose and feel
resentment
5 Work through the differences in a frank and honest
way with mutual trust and respect. This is essentially
the ideal response from an effective team and the
principal way of diminishing hostility and conflict,
but unless one has trust and respect it is difficult to
implement; which comes first the chicken or the egg?
Finally, one of the principal reasons for conflict is a
history of previous hostility and conflict. At any one point in
time there may be ill feeling between groups in a company
which has built up over the years and which is very difficult
to resolve within the temporary lifespan of a project. This
may be the project manager's own company or it may be one
company in the global organisation. There is generally a
new global organisation for every project, and any one of the
companies involved may have this situation, which if not
resolved for the individual project, can spread through the
total organisation like a disease.
The resolution of conflict
One of the project manager's principal tasks is that of an
integrator and as such, the management, prevention and
resolution of conflict is his responsibility. It is often
impossible to eliminate differences, disagreements and
354
competition completely, and it is questionable if it is

advisable to do so. An organisation without visible signs of
these natural human traits is normally mediocre, overly
conformist and a dull place to work in. These factors
contribute to better performance, problem solving, innovation, decision making and commitment to objectives, provided they are constructive. If they are about how best to
carry out the project, then they are beneficial to the project It
is when they degenerate, and the project is not so important
as scoring over the opposition, that conflict is harmful to
achieving the project objectives.
A method of resolving differences, without creating
conflict, is not to deny that they exist, but to accept that they
occur naturally and to endeavour to resolve them by the
logic of the situation. This is not just appealing to the groups
to be objective, but also to take into account the more nontangible contributors to conflict. If there can be frank and
open discussion, that is, levelling, and a mutual analysis of
the problem and the solution with respect for the other's
point of view, then differences can be worked through and
resolved without adverse conflict. This can lead to a greater
understanding and commitment to the outcome, and if
successful contributes to the growth of an effective team.
Unfortunately the working through of differences in this
manner is one of the attributes of an effective team, and
without having the makings of an effective team it is difficult
to achieve; a chicken and egg problem again. Thus the
problem the project manager is faced with is how to create
the conditions, and how to provide the leadership and
management that will increase the chances of an effective
team evolving and prevent or reduce the likelihood of
adverse conflict arising.
The various steps he can take can be classified into three
sets, namely:
1 Formal organisational steps.
2 Informal organisational steps.
3 Managerial or personal behaviour steps.
The biggest advantage the project manager has is, as before,
that there exist what the behavioural scientists call a super-
355
ordinate objective, that is, the project itself. It has been

shown that the biggest factor leading to a reduction of
conflict is for all the groups to have a common objective to
which they are committed and that requires them to interact
with each other. Thus if the project attitude can be established,
the project manager is halfway or more to defusing adverse
conflict. In addition, project work is normally challenging
work, and when individuals are involved in challenging
work there is less likelihood of conflict.
To encourage the establishment of this superordinate
objective, all the organisations involved must emphasise
that the criteria of individual and group performance is the
total project organisation's effectiveness, and not that of the
individual or single group. This involves such factors as
performance assessments and salary increases and promotions being based on overall project performance, and thus
the project manager having a say in these factors. It also
involves the companies concerned recognising that the
global project organisation is an entity which exists, and
giving it due recognition.
To ensure that commitment to the project is built up and
maintained the project manager must have a formal and
informal information system that keeps everyone involved
informed of progress and permits open communication.
The project objectives, progress, problems and success must
be communicated to everyone involved. There must also be
as clear a definition as is possible of the responsibilities of
those involved. The use of the work breakdown structure,
cost accounts, work packages and matrix of organisation
charts enables individuals right down the line to know what
these responsibilities are, who they must interact with and
the role of the project manager.
Not only must formal organisational factors be used to
encourage the formation of an effective mixed project group,
but so should informal factors. This implies encouraging the
formation of informal group and social interaction between
groups, for example, involvement in sports, having dinners
together, dances, cocktail parties and having coffee and
lunch in the same group. Physical factors should be used to
stimulate project groups; for example, a common office for
356
ADVANCED PROECT MANAGEMENT
those involved, partitioned areas instead of a large 'bullpen', visits to each other's companies. It is not generally
possible to form an effective team with people you don't
know fairly well, and thus both the owner and contractor
should encourage prolonged visits of key personnel to each
other's offices. Once people have had face to face contact
and know each other, physical proximity is not essential in
the long term. Without this, communication will be limited
to formal channels, and without informal information
channels, effective communication between groups will be
greatly handicapped.
In dealing with the people and groups involved in his
project, the project manager must be aware of the likelihood
and hazards of conflict, and use tact in his dealings. Though
he must of necessity use pressure, he must be aware that if
pushed too hard into a corner, an individual can only resist,
and the harder the project manager pushes, the harder the
individual will push back Thereafter he must show respect
for and listen to opposing points of view, and on occasions
he may have to compromise and back down on a point, in a
conscious effort to manage conflict He cannot afford to
become too emotionally involved and he must at all times
keep calm. One of the basic lessons of management is that if
you cannot manage your own emotions in the workplace,
you cannot manage other people.
The project manager must be aware of the problems
involved in the other functions involved in the project, and
this often involves rotation of personnel among the basic
functions. The project manager who has worked for both an
owner and a contractor company will be in a better position
to appreciate the other 'side's' problems and position.
Finally, the project manager must be aware that in a matrix
organisation he is impinging on other managers' roles,
authority and territory, and he must do so with caution,
respect and tact. Given all that, he must still attempt to
minimise time and cost on his project and he is not in a
popularity contest. At times he must throw caution to the
winds and go 'bull headed' for what he believes to be
necessary, but he must also be aware of the consequences
and the alternatives. An effective team will respond to a
HUMAN BEHAVIOUR IN THE PROJECT SETITNG
357
greater challenge with more commitment to succeed, and he

may be able to achieve his objectives without creating
conflict.
APPENDIX: Planning Techniques

Planning techniques are of two main types, namely bar
charts and the critical path method. In both of these planning techniques the planning data base, including the
relationship between activities or jobs, is required as shown
in Figure Al. In bar charting the relationships are used
explicitly in the construction of the plan, but are only
implicit in the final plan. In the critical path method the
relationships are explicit at all times.
of Men
Estimate
(~000)
No.
No.
Activity
Time
Predecessor
activities
Basic d e s i g n
16
154
Aechanical d e s i g n I
12
58
Basic d e s i g n
16
17
Basic design
None
Purchase mechanical
parts
565
Mechanical d e s i g n I
Mechanical d e s i g n I1
12
86
Mechanical design I
Specify e l e c t r i c a l
motors
19
Purchase e l e c t r i c a l
parts
3 60
E l e c t r i c a l d e s i g n I1
2
8
16
154
Instrument design
12
115
10
Purchase instrument9
270
11
D e l i v e r mechanical
parts
Purchase mechanical
97
Mechanical design I
12
Mechanical
drawing o f f i c e
16
Figure A1 Planning data base
arts
359
APPENDIX
NO.
Estimate
of Men
(~000)
Predecessor
activities
NO.
Activity
13
motor
Specify e l e c t r i c a l
parts
Deliver e l e c t r i c a l
pasts
parts
E l e c t r i c a l drawing
office
Instrument drawing
office
Deliver instruments
Fabricate mechanical
parts
Deliver motors
Fabricate e l e c t r i c a l
parts
Assemble mech I
Assemble e l e c I
Assemble mech I1
Assemble e l e c I1
Time
Instrument design
Purchase instruments
Deliver mechanical
parts
motors
Deliver e l e c t r i c a l
parts, electrical
drawing off i c e
F a b r i c a t e mech p a r t s
deliver electrical
motor
Fabricate e l e c t r i c a l
p a r t s , instrument
drawing office.
Deliver instruments
Assemble mech I
Assemble e l e c I
F i n a l assembly I
Assemble mech I
Assemble e l e c I
F i n a l assembly I1
Assemble e l e c I
Assemble mech 11
Final assembly I
Figure A1 (concluded)
Bar Charts
A bar chart consists of a sheet of paper, sometimes graph

paper, with the activities making up the project listed down
the left hand edge and with a horizontal time scale. The
activities are generally listed from the top down in the
general sequence in which they are expected to be carried
out, within each major segment or group of activities. For
example, the activities involved in mechanical sub-assembly
360
APPENDIX
will be grouped together as will the electrical sub-assembly

and final assembly.
Once this format is laid out, the work on each activity is
shown by a line or bar on the horizontal scale opposite the
activity listing. The length of the bar is scaled to be
equivalent to the time necessary to complete the activity and
its position on the horizontal time scale represents the time
at which it should start and finish. Thus the sequencing and
interrelationship shown on the data base is used to place
each bar in the position or time it can be carried out.
For example, the first activity is basic design, so it is scaled
off as shown in Figure 5.8. Once this is completed, mechanical design I and electrical design I can be scaled off on the
chart. After mechanical design I is scaled off, purchase
mechanical parts and mechanical design I1 can be charted
and so on for all the activities, Thus working from left to
right, that is from start to finish, all the activities can be
charted.
This gives the first iteration of the project plan, but as
previously discussed work planning is not completed until
the manpower is planned and the same bar chart can be
used for this. This is shown in Figure 5.8 by a single number
above each bar representing the number of men estimated to
complete each activity in the time shown. In practice each
trade or profession can be shown thus using suitable coding
and a separate assessment of manpower requirements made
for each.
For simplicity only one figure is used in this example.
It is then relatively simple to determine the manpower
required for the project by summing up for each time period
the manpower shown on each activity taking place in that
time period. For example, in week 20:
16 men are required for electrical design II
12 men are required for instrument design
16 men for the mechanical drawing office
4 men for the purchase of electrical motors
48 total manpower required in week 20
The total manpower and the manpower for critical trades
can then be shown as a graph or histogram as in Figures 5.2
and 5.3. They can also be shown as a graph of the cumulative
APPENDIX
361
manpower, which is the basis of an effective control tool, the

S chart.
The same method can be used to produce a time-phased
budget by noting expenditure on each bar in a similarway to
manpower and summary for each period.
The critical path method

In this method of planning the planner takes the base data
and the relationship between activities and:
1 Constructs an arrow diagram, sometimes termed a

network, showing diagrammatically the sequence
and interrelationship of activities, i.e. the planning
'logic', as shown in Figure 5.9.
2 Carries out a time analysis which shows the time
framework within which activities must be carried
out.
These two steps constitute the basic form of the critical path
method, but in practical applications two further steps are
necessary:
3 A resource analysis is carried out, i.e. manpower

planning to balance the resources available with the
time analysis.
4 The time phased budget is prepared.
The definitions of the terms used are shown in Figure A2.
Whereas in the bar chart method the project plan consists
of a map-like chart, in the critical path method the project
plan is represented by the arrow diagram and the time
analysis, as shown in Figure 5.9 and 5.10. The time analysis
shows for each activity:
Its earliest start time
Its latest start time
Its earliest finish time
Its latest finish time
Its independent float
6 Its free float
7 Its total float
Those activities with zero float are on the critical path.
1
2
3
4
5
362
APPENDIX
Network analysis definitions
- is an element
An a c t i v i t y
An event
of t h e work e n t a i l e d on t h e p r o j e c t : an
a c t u a l job o r task: e.g. waiting f o r d e l i v e w of e q u i w e n t ;
o b t a i n approval fmm board.
is the start o r f i n i s h of a group of a t i v i t i e s o r s i n g l e

a c t i v i t y ; it i s a p o i n t i n time.
The e a r l i e s t s t a r t time
The l a t e s t f i n i s h time
The l a t e s t s t a r t time
Independent f l o a t
Free f l o a t
Total f l o a t
p o s s i b l e time a t which an
a c t i v i t y can s t a r t ( e a r l i e s t s t a r t time of t a i l
event).
is t h e l a t e s t event time of t h e head event.
- is
t h e l a t e s t p o s s i b l e time by which an a c t i v i t y
can start without holding up t h e t o t a l project.
is t h e time by which an a c t i v i t y can expand without

a f f e c t i n g any o t h e r previous o r subseollent w e n t .
i s t h e time an a c t i v i t y can expand without a f f e c t i n g

subsequent a c t i v i t i e s .
C r i t i c a l path
- i6 t h e e a r l i e s t
i s t h e time by which an a c t i v i t y can expand without

affecting- o v e r a l l p r o j e c t time (reduces f l o a t on previous
and subsequent a c t i v i t i e s ) .
- is
t h a t sequence of a c t i v i t i e s which determines t h e t o t a l

time f o r t h e p r o j e c t . The sequence o r path extends from
t h e s t a r t event r i g h t through t h e p r o j e c t t o t h e end w e n t
and i t i s t h e sequence of a c t i v i t i e s with t h e longest
totalltime. There may be mare than one c r i t i c a l path i n q v
project.
Figure A2 Definitions in critical path methodhetwork

analysis
Constructing an arrow diagram
The arrow diagram is constructed as follows, and as illustrated in Figure 5.9.
1 Each activity is shown as a n arrow, with the tail ofthe
arrow denoting the start of the activity and the head
the finish.
2 Each activity, i.e. arrow, has an abbreviated description above it.
3 Each activity has the time required to complete it
noted below the arrow.
4 The events at the tail and head of the arrow, i.e.
activity, are shown as circles, or, in more sophisticated
diagrams, as rectangles with spaces for event data
within them.
APPENDIX
363
5 Activities are linked together by events to form the

sequence of activities and this makes up the arrow
diagram.
6 Events are given a unique number, conventionally in
ascending order from start to finish of the project.
(This is a necessity in some, generally older, computer packages.)
7 Each activity is thus uniquely defined by the number
of the event and its head and tail, i.e. the head and tail
event number. Conventionally the tail event number
is termed I and the head event number termed J.
8 When two or more activities have the same head and
tail event, a 'dummy' activity with a zero time
duration is added together with another event, as
shown in Figure A3.
9 Dummy events are required in two other situations
to illustrate particular interrelationships which
sometimes occur (see Figure A3).
(a) Logic dummy
(b) Ladder dummy
Carrying out the time analysis
Once the arrow diagram is constructed and the time noted

on it, a time analysis can be carried out on these stages:
1 The earliest event times are determined: the forward
pass
2 The latest event times are determined: the backward
pass
3 The activity time output, i.e. earliest start etc., is
calculated using the earliest and latest event times.
Determining the earliest event times
1 The earliest event time of the start event of the project

is set at zero, and is conventionally shown above the
event circle.
2 The earliest event times of succeeding events are
calculated by starting at the first event, adding the
364
APPENDIX
(a) Two or more activities have the same head and tail
event
(7'
%@ )
R
I Dummy
15
Incorrect
-1 -
Correct
-
(b) Logic Dummy

Activity C has predecessors A and B
Activity D has predecessor B
/ \
Dummy
Correct
Incorrect
(c)
Ladder Dummy
When two or more activities overlap, such as the manufacture of sub-assemblies and final assembly, a dummy
or dummies is used to represent this. For example,
when an initial batch of sub-assemblies is completed,
work can start on the final assembly, and when subassembly work is completed there is a period of subassembly. This is handled as shown below.
Sub. A s s . 1
Sub. A s s . 2
Figure A3 Dummies
APPENDIX
365
activity time to the earliest event time of the activity

tail event. Thus starting at the start event, the earliest
event times can be calculated by a forward pass of
calculations.
3 A problem occurs when two or more activities finish
at the same event. The earliest event time forthis type
of event has to be calculated for each sequence of
activities or path leading to it, and the highest earliest
event time so calculated is then used.
4 When the last event is reached, the earliest event time
of that event is the earliest project completion time.
Determining the latest event time
1 The latest event time of the last event is set as the

same as its earliest event time, and is conventionally
shown below the event circle.
2 The latest event time of preceding events are then
determined by starting at the last event time and
subtracting the activity time from the latest event
time of the head event, i.e. the backward pass of
calculations.
3 Aproblem occurs when two or more activities start at
the same event. The latest time for this type of event
has to be calculated for each sequence of activities
leading to it from the last event. The lowest event
time so calculated is then used.
4 When the first event is reached, the latest event time
of that event should be zero.
If an event has the same earliest and latest event
time it is on the critical path.
Canying out the time analysis activities
The event times calculated by the above methods are then

used to calculate the time framework for the activities, that is
the basic planning output of the critical path method. Two
conventions are used which slightly change the calculation
rules:
366
APPENDIX
1 The earliest start time of the first activity is set at time

zero.
2 The earliest start time of the first activity is set at day
or week number, i.e. the dating convention.
As the first convention is used in this book, the calculation
rules will be first outlined for that convention, and the modifications to these rules noted later.
The calculation rules for the activity time analysis are:
1 The earliest start time of an activity is the earliest
event time of its tail event.
2 The latest start time of an activity is determined by
subtracting the duration time of the activity from the
latest event time of the head event.
3 The earliest finish time of an activity is the earliest
event time of the tail event plus the activity time.
4 The latest finish time of an activity is the latest event
time of the head event.
5 The independent float of an activity is the difference
between the earliest event time of the head event and
the latest event time of the tail event, less the activity
time.
6 The free float of an activity is the difference between
the earliest event times of the head and tail events,
less the activity time.
7 The total float of an activity is the difference between
the earliest time of the tail event and the latest time of
the head event, less the activity time.
When the second convention, that is the dating convention,
is used the first two rules for determining the start times of
the activity are modified as follows:
1 The earliest start time of an activity is the earliest
event time of the tail event plus one.
2 The latest start time of an activity is determined by
subtracting the duration time of the activity from the
latest event time of the head event, and adding one.
As is implied by its title, the dating convention is normally

used when actual dates are used in the time analysis output
APPENDIX
367
That is, there will be a table used to convert day or week

numbers to actual dates.
Manpower planning
Unless a computer is used for resource analysis, manpower

planning is carried out by constructing a bar chart from the
time analysis output. On this type of bar chart the activity
bars are placed at the earliest start time of the activity and the
total float is shown as a dotted extension of the bar. Figure
5.12 shows such a bar chart in one of the forms produced by a
computer.
Time limited resource analysis allows you to move activities within this float to reduce the peak demands for
manpower. However, if this does not reduce the peak
demand to that available, then that peak demand must be
accepted. Resource limited analysis allows you to move
activities more than this float so that peak demand never
exceeds the manpower available. However, this means that
the project time will be extended.
Bibliography
Avot, I., 'Why does Project Management fail?, Management

Review, Jan. 1970, pp 36-41
Avot, I., 'Cost-relevance Analysis for Overrun Control',
Project Management, Vol. 1, No. 3, August 1983, pp 14248
Baumgartner, J. S., (1964), Project Management, Irwin
Baumgartner, J. S., (1979),Systems Management, The Bureau
of National Affairs, USA
Cleland, D. I., 'Organisational Dynamics of Project Management', E E E Transactions in Engineering Management.
Feb. 1969, pp 96-106
Cleland, D. I. and King, W. R, (1968), Systems Management
and Project Management, McGraw-Hill
Cost/Schedule Control Systems Criteria, Joint Implementation
Guide, (1976), Departments of the Air Force, the Army,
the Navy and the Defense Supply Agency, USA
Hackney, J. W., (1977), Control and Management of Capital
Projects, McGraw-Hill
Handy, C. B., (1979), Understanding Organkations, Penguin
Books
Hanson, J. T., 'The Case of the Precarious Program', Harvard
Business Review, Jan.-Feb. 1968, pp 14-34, 170
Jack, k B., Purchasing in a Heavy Engineering Zndusm,
Anderson Strathclyde Ltd., an internal report
Jennet, E., 'Experience with an Evaluation of Critical Path
Methods', Chemical Engineering, Feb. 1969, pp 46-106
Kharbanda, 0. P., et a1 (1980), Project Cost Control in Action.
Gower
BIBLIOGRAPHY
369
Knight, K., (1977), Matrix Management, Gower

Krause, W. A., 'Cost Control: A Contractor's Viewpoint',
Chemical Engineering Progress, Dec. 1968, pp 15-19
Lock, D., (1984), Project Management, 3rd ed., Gower
Lockyer, K G., (1975), An Introduction to Critical Path
Analysis, Pitman
Lockyer, K G., (1966), Cntical Path Analysis: Problems and
Solutions. Pitman
Management Contracting, Construction Industry Research
and Information Association, CIRLA Report 100
McGregor, D., (1960), The Human Side of Enterprise,
McGraw-Hill
McGregor, D., (1967), The Professional Manager, McGrawHill
~ c ~ h a N.
i l A,
, (1980), The Engineer as a Manager, Unpublished M. Eng. Dissertation, University of Glasgow
Marcia riello, J. A, (1978), Program Management Control
Systems, J. Wiley and Sons
Middleton, C. J., 'How to Set up a Project Organisation',
Harvard Business Review, Mar.-Apr. 1967, pp 73-82
McNeil, J. F., Program Control Systems, IEEE Transactions
in Engineering Management, March 1964
Militaly Standard Work Breakdown Structures for Defense
Material Items, (1975) Department of Defense, USA
North Sea Costs Escalation Study, (1976), Department of
Energy, Peat Marwick Mitchell & Co., and Atkins
Planning, HMSO
O'Brien, J. (ed.)., (1969), SchedulingHandbook, McGraw-Hill
Peart, A. T.,(197 l), Design of Pro~ectManagementSystems and
Records, Gower
Perry, J. G. and Thompson, P. A., Target and Cost Reimbursable Construction Contracts, CIRTA Report 85
Pilcher, R, (1973), Appraisal and Control of Project Costs,
McGraw-Hill
Proceedings of Internet 76, Project Implementation and
Management: Bridging the Gap, 5th Internet World
Conference, Vols. 1-5
Proceedings of the 6th Internet Congress, 1979, Vols. 1-4
The Project Manager, Journal of the Association of Project
Managers
370
BIBLIOGRAPHY
Saynisch, M., 'Project Management System for a Large

International Project', Project Management, Vol. 1, No. 2,
May 1983, pp 115-21
Schein, E., (1970), Organisational Psychology, Prentice Hall
Taylor, W. J. and Watling, T. K, (1979), Successful Project
Management, Business Books
'Time and Resource Aspects of Project Management in the
Construction of Chemical Plants', The Chemical
Engineer, June 1967
Index
Actual cost ofwork performed, 249,251

ACWP, 249. 251
Advanced project management
elements of, 203
Art of planning, 22
Assessment of suppliers, 308
Attitude, projecr 338
Authority problems in project management, 322
influence on form of organisation
used 327
definition of, 323
fonnal, 324
informal, 324
in the matrix organisation. 327
the plan as a basis of, 326
Bar charts
advantages and disadvantages of,
.136
..
in hierarchical planning, 221

in modem project planningmethods
154
use with CPM/PERT. 149
versus CPM/PERT, 141
BCWP, 249
BCWS, 249
Bidlist, 112
Budgeting
dynamic approach to. 131
in cash flow forecasting. 134
in launch planning 30
time phased as a basis for control,
126
Budgeted cost of work performed, 249
Budgeted cost of work scheduled, 249
Buying 304
Cash flow forecasting, 134

Changes
control of. 275
detrimental effect of, 276
effect on escalation of cost, 82
endemic in the larger prnjecl 199.
200
reasons for. 277
requirements for control of, 279
Change control systems
baselines for control of change, 282
control board 285
description of, 280
identifying, 283
integration with other systems. 287
request form. 285
Coding. 270
Computers
based systems, 289
bureaux systems, 298
company systems. 297
dedicated to project management,
299
in performance analysis, 263
in Integrated Project Management
Information System. 292
materials managemenr 294
micro-computers, 299
mini-computers, 300
modem computerbased techniques
of planning 154
problems within planning, 147
project planning packages. 152
records management 294
Concept of project management 8
Configuration control, 280
Conflict
consequences of. 348
INDEX
Conflict conlinued
intergroup. 347
in the matrix organisation, 351
methods used to resolve. 353
reasons for, 350
resolution of. 354
Contingency. 87
Conhngency versus job-schedule, 94
Contract
cost plus 107
form of, 103
influence on cost control. 104
influence on cost escalation, 103
lump sum, 105
other forms, 109
Control
and variance analysis, 65
baselines for. 126
by involvement, 58
centres, 241
cycle, 31, 55
forward looking 68
functions of, 55
inadequacy of historical accounting
methods, 67
informal system$ 61
integration of sub-systems for, 63
introduction to, 55
managerial function of, 57
managerial philosophy of. 57
modem methods of, 241
motivational function of, 275
of changes, 275
of costs. 62
of the larger project, 202
planning, 31
procurement. 302
quality of control and information
systems. 71
real time, 59
responsibility for, 58
requirements of a good control
system, 76
subjectivity in, 75, 246
why ineffective. 60
Co-ordinator, project, 11.327
Cost
accounting 62
consciousness, 98
consciousness in design, 94
control, 62
definition of cost control 62
design to, 102
forecasting final, 261

performance index 260
variance analysis of. 254
Cost accounts
as a pseudo sub-contract. 219
as a small project 220
definition of. 21 1
hierarchy of. 217
plan. 226
report. 271
CPMIPERT
advantages of. 141
float. management of, 170
float, problems with. 146
problems with, 144
communication, 144
computer-based systems. 147
control 145
float, 146
human factors, 143
operational planning. 145, 171
sorted outpuf 156
use with bar charts, 149
versus bar charts I41
Critical path method. see CPMPERT
Data overload 268

Design
cost consciousness in, 94.98
optirnisation, 102
planning, 47
standards, 101
to cost. 102
Detail. level of. 35
~ivisibnalproject organisation I3
Dual subordination 321
EAC, 249
Earned value. 248
Engineer
as a manager, 329
career path, 331
as a functional manager. 332
as a general or project manager, 332
Escalation
clause. I14
definition of, 81
factors causing 82
Estimate at completion, 249
Estimate to complete, 249
Estimating
characteristic flow of inrormation,
86
evolution of, 88
importance of, 79
stages in. 88
the-poject costs, 79
ETC, 249
Exoeditine
a n 4 306
effective, 3 11
Function of control, 55
Function of planning, 21.28
Global project organisation, 17
Groups
horizontal, 340
in the project setting 339
mixed, 340
vertical, 340
Hierarchical planning, 221
Human behaviour in the pmject setting,
319
In-company project organisation, 11
Inflation
consequences of. 82
effect on escalation of costs, 84
forecasting, 83
project costs, 82
Influence, 324
Interpersonal behaviour, 333
Larger project, I98
Level of detail in planning 35
Levels in planning, 221
Lines and staff forms in project organisation 12
Line of balance, 189
LOB, 189
Management inrnrmntion uyutems. 289
Management philosophy ofconrrol. 57
Management ph~losophv
plannine,
. . of .
24
Matrix of responsibilities 21 I
Matrix organisation 14
authority problems of, 327
complexities of, 321
Milestones, 169
Network analysis see CPMIPERT
Operational planning
problems with. 145
modem methods of. 171
Organisation
co-ordination form of, 11
divisional 13
global, 17
in-company, I I
matrix, 14
of projects, 11.29
temporary nature of. 321
Overheads 244
Participation, 336
Performance analysis, 241
Performance report, 271
PERT. see CPMIPERT
Plan as a management tool, 33
Plan as a tool to achieve authority, 326
Planning
art of, 22
basic steps in, 1I6
completion and commissioning, 53
computer-based systems. 150
contribution oS 21
control 31
data base for, 154
design, 47
detail, level of, 35
function of, 28
human factors in. 22
hierarchy of plans, 221
larger project, the, 198
launch. 28
levels of planning, 221
library modules, 183
line of balance, 189
LOB. 189
manoger~alphilosophy. 24
manpower. I21
mrlestones. 169
modem methods of. 154
INDEX
Planning continued
multi-project planning, 230
operational 145, 171
planning the planning process, 38
portfolio of projects. 198. 230
rolling wave concept 227
reports, 154
resistance to, 25
resources. 174
S curves, 157
science of, 120
sorted output I56
specialist 25
the control and information
systems, 30
the management, 29
very large project 236
Power, 324
Project
attitudes. 338
complexity of, 4
definition of, 1
objectives, 40
strategy, 40
Procurement
buying 304
planning and control of 302
project liaison, 303
suppliers' plans. 314
Real time control, 59
Real time planning, 34
Reports
higher level 271
performance, 257
Resistance to planning. 25
Role of stan-planner, 27
Schedule performance index, 260
Schedule variance. 254
S curves, 157
Shortcomings of project management
10
Sorted output, 156
Split responsibility. 320
Standards in design. 101
Suppliers
assessment of, 308
plans, 314
Systems
integration of, 29,292.289
project management information,
289
Target completion dates, 41
Team development 344
Temporary nature of projects. 321
Traditional management theory. 5
Variance analysis
traditional, 65
costs, 254
schedule. 254
Why project management?, 3

Work breakdown structure, 204
Work packages, 242

Advanced Project Management - 3ed Edition

Uploaded by

Copyright:

Available Formats

Advanced Project Management - 3ed Edition

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Advanced Project Management - 3ed Edition

Uploaded by

Copyright:

Available Formats

All rights reserved. No part of this publication may be

British Library Cataloguing in Publication Data

Typeset in Great Britain by

Estimating the Project Cost

Planning the Smaller Project and the Basic Tools 116

The Planning and Control of the Larger Project, or

The Elements of Advanced Project Management 203

Project Management Computer-based Systems

Planning and Control of Procurement

Human Behaviour in the Project Setting

Appendix: Planning Techniques

Contribution of project planning

Effect of inflation on project expenditure

Multi-stage explosion of project

Project plan for simple example

Integrated project management information

Planning data base

Preface to Second Edition

Project management is a highly professional branch of

a professional level guide to the management, planning

Many people drift into managing or co-ordinating projects

PREFACE TO SECOND EDITION

or problems involved. They often perform satisfactorily on

PREFACE TO SECOND EDITION

industries, experience with modern computer-based

PREFACE TO SECOND EDITION

larger project apply to the management of a portfolio of

PREFACE TO SECOND EDlTION

company project organisations. The following chapters deal

Many people have contributed, directly or indirectly to this

PREFACE TO SECOND EDITION

disappearing for long hours in the evening or at the

A project can be defined as a non-routine, non-repetitive,

ADVANCED PROJECT MANAGEMENT

patterns, and specialist techniques are necessary. Other

problems, planning and control techniques are almost

Why project management?

ADVANCED PROJECT MANAGEMENT

planning financial management control and the particular

which in turn compounds these problems. Great em~ha@s

exp~nditu~6e~use~of~inadequ~te~plan-g~and~controlsystems. Shecia1 project-oriented information systems are

Traditional management theory and project management

ADVANCED PROJECT MANAGEMENT

ot handle projects efficiently, tends not to meet schedules

ADVANCED PROJECT MANAGEMENT

need for a separate management function to manage projects.

drive necessary to ensure completion on time and within

ADVANCED PROJECT MANAGEMENT

and the human relation skills required. These functions are

1 The general management and logistics problems

The form of organisation used to manage projects is often

organisation structure be related to the work to be done on

In-company project organisation

1 A line and staff organisation, i.e. a project coordinator.

The first stage of evolution to some form of project-oriented

ADVANCED PROJECT MANAGEMENT

tional line and staff functional organisation essentially