Logistics

Lecture notes

Maria Grazia Scutell

Dipartimento di Informatica
Universit di Pisa

September 2015

These notes are related to the course of Logistics held by the author at the University of Pisa.
They are based on some books from the literature, as acknowledged at the end of each chapter.
Chapter 1

Introduction to logistics systems

Logistics deals with planning and control of material flows and related information, in
public and private organizations.

A relevant issue in Logistics is to take decisions (e.g. how and when raw materials should
be acquired), by satisfying a given set of constraints (e.g. a budget constraint) while
optimizing a certain performance measure (e.g. minimizing the total cost).

Logistics plays a great role in three main contexts:

military context, i.e. supply of troops with food, armaments and troop transporta-
tion;

civil organizations, i.e. production and distribution in firms;

public organizations, i.e. service management (e.g. garbage collection, mail deliv-
ery. . . ).

In these notes the emphasis will be on logistics systems which are typical of civil or
public organizations, where they have a very significant impact. For example, the total
logistics cost incurred by USA organizations in 1997 was higher than the overall USA
expenditure in social security, health services and defence.

Specifically, a logistics system is made up of a set of facilities linked by transportation

services, where:

facilities are sites where materials are processed (produced, stored, sold, consumed. . . );
they include manufacturing centres, warehouses, distribution centres (DC), trans-
portation terminals. . . ;

transportation services denote the movement of materials between facilities (using ve-
hicles and equipment), and they are usually depicted as directed arcs:

7
8 1.1. Supply chains

facility facility

1.1 Supply chains

The set of facilities and transportation services is called supply chain. It models the
complex logistics system where raw materials are converted into final products and then
distributed to the final users, by considering a business or a service activity context, as
indicated before.

Figure 1.1 illustrates a typical supply chain where production and distribution are made
up of two stages each.

product flows

manifacturing
supplier plant
assembly RDC
plant CDC
RDC

RDC
supplier manifacturing CDC
plant RDC

outlets
production system distribution system

information flows

Figure 1.1: A typical two-stage supply chain. CDC: Central Distribution Centres;
RDC: Regional Distribution Centres.

This is a very general and abstract representation; usually each facility ( ) may com-
prise devices and subsystems (machines, retrieval systems. . . ), while each transportation
link ( ) may denote a simple transportation line (e.g. a truck line) or a more com-
plex system. In other words, the exact meaning of facility and transportation link does
depend on the level of detail we want to address.

The supply chains can be classified according to different criteria.

Chapter 1. Introduction to logistics systems 9

Push vs. Pull supply chains

pull or make-to-order (MTO): in this kind of supply chains, finished products are man-
ufactured only when customers need them, and so inventories are not maintained;
push or make-to-stock (MTS): in these supply chains, production and distribution
decisions are based on forecast; therefore, inventories can be managed;
mixed approaches such as make-to-assembly (MSA) supply chains are also possible,
where the final assembly stage is pull based.

Degree of vertical integration and third-party logistics

vertically integrated : in this case all the supply chain components belong to a
single firm (quite rare);
supply chains are operated by several independent companies (more frequent).
In the second case, the relationship among companies can be:
transaction based and function specific;
strategic alliance; examples are given by:
third-party logistics (3PL): an outside company performs all or part of a
product distribution;
vendor-managed resupply.
The main logistics activities and the related decision problems will be addressed in the
rest of the notes.

1.2 Logistics managerial issues

When devising a logistics strategy and taking decisions, managers typically try to
achieve a compromise among three main objectives:
1. capital reduction: to reduce as much as possible the level of investment in the
logistics system (logistics network, equipment. . . );
2. cost reduction: to minimize the total cost associated with operating the supply
chain (inventory management, transportation. . . );
3. service level improvement: this is also a relevant objective, since the level of logis-
tics service influences customer satisfaction, which in turn has impact on revenues;
often it is expressed via the order-cycle-time, i.e. the elapsed time between the
instant an order (or a service request) is issued and the instant foods are received
(or service is provided).
10 1.3. Decisions in logistics systems

Since the three objectives may be contrasting, often a customer service level is set (first
phase), then logistics decisions are taken so as to meet that service level at a minimum
(capital and operating) cost.

1.3 Decisions in logistics systems

Several decisions have to be taken in logistics systems, both at the design and at the
operating level. Examples are:
location problems: should new facilities (e.g. CDC, RDC. . . ) be opened? what
are their best configuration, size and location?
production problems: how should production be planned?
inventory problems: when and how should each stocking point be resupplied?
transportation problems: what is the best fleet size? how should vehicles be
routed?
Logistics decisions are traditionally classified as follows, according to the considered
planning horizon:
strategic decisions: they have long-lasting effects (many years), and include logis-
tics system design and acquisition of costly resources (facility location, capacity
sizing, plant layout, fleet sizing); they often use forecasts based on aggregate data;
tactical decisions: these decisions are made on a medium-term basis (e.g. monthly);
they include production and distribution planning, inventory management, stor-
age allocation, transportation mode selection. . . ; they often use forecasts based
on disaggregate data;
operational decisions: in this case, decisions are made on a daily basis or in real-
time; they include shipment and vehicle dispatching, vehicle routing and schedul-
ing, and are based on very detailed data.

1.3.1 Methods for making decisions

How can we take logistics decisions?

Quantitative analysis is essential for intelligent logistics decision making. Operations
Research offers many support methods for taking decisions in Logistics. Three basic
methods are:
1. Benchmarking: this method is helpful when a logistics system already exists, and
one wants to compare its performance to a best practice standard, i.e. the one of
an industry leader in logistics operations; the most popular method is based on the
Supply Chain Operations References (SCOR) model, using several performance
indicators.
Chapter 1. Introduction to logistics systems 11

2. Simulation: this methodology is helpful to answer a number of what-if questions

regarding different alternatives to the existing system, by considering the dynamics
of the system (e.g. what is the average order retrieval time if we use a certain
policy storage rather than the existing one); however, this approach may be time
consuming and heavy in case of a large number of alternatives.
3. Optimization: some logistics decision problems can sometimes be formulated as
mathematical optimization models. This is the approach which will be pursued
in these notes. Optimization models include:
Linear Programming (LP) models
Network Flow models;
these are easy problems, i.e. solvable in polynomial time. However, most logistics
problems can be formulated only in terms of
MixedInteger Linear Problems (MILP)
Nonlinear Problems (NLP);
they can be difficult to optimize, since NP -hard. This has motivated the devel-
opment of heuristic approaches, in order to find good but not necessarily optimal
solutions.
In any case, modeling logistics problems in terms of LPs and MILPs is very important,
both to use suitable solvers to get solutions (for easy problems or difficult ones of
reasonable size), or as a starting point to design exact and/or heuristic approaches for
difficult problems. This is the subject of these Logistics notes, with focus on network
flow, LP and MILP formulations to state and solve Logistics decisions problems.

References G. Ghiani, G. Laporte, and R. Musmanno (2004): Chapter 1

Textbooks

Bigi, G., A. Frangioni, G. Gallo, and M.G. Scutell (2014). Appunti di Ricerca Op-
erativa. Italian. url: http : / / didawiki . di . unipi . it / doku . php / ingegneria /
ricercaoperativa1/start#testi_di_riferimento.
Drezner, Z. and H.W. Hamacher (2004). Facility Location: Applications and Theory.
Springer series in operations research. Springer. isbn: 9783540213451.
Ghiani, G., G. Laporte, and R. Musmanno (2004). Introduction to Logistics Systems
Planning and Control. Wiley Interscience Series in Systems and Optimization. Wiley.
isbn: 9780470091654.
Ragsdale, C.T. (2004). Spreadsheet Modeling and Decision Analysis: A Practical Intro-
duction to Management Science. Cengage South-Western. isbn: 9780324321777.
Toth, P. and D. Vigo (2002). The Vehicle Routing Problem. Monographs on Discrete
Mathematics and Applications. Society for Industrial and Applied Mathematics. isbn:
9780898715798.

123