White Book Mancini

Advances in plant
modularisation
From the state of art
to emerging challenges
Edited by Mauro Mancini

ISBN 9788888198361
© ANIMP SERVIZI srl

May 2014, Milan - Italy
with the support of

Preface
I’m proud to introduce this publication that could be the first volume of a series aimed at
boosting innovation and at capitalizing lessons learned in the Engineering and Construction
Sector. Industrial plant modularisation represents one of the main drivers for the innovation
and competitiveness of both clients and main contractors in the future decades. Modularisation
is the basis of power plant flexibility (both in the management of large power plant fleets and in
the ordinary single plant exercise) that will become the real turning point for the new
generations of industrial plant engineering and management approaches.
ANIMP’s main mission is to support the sharing of Industrial Plant culture all over the world
through its technical chapters that merge the experiences of the Associated Companies in the
specialist disciplines. The over cited collaboration between Industrial Companies and
Universities in the industrial plant engineering and management is efficiently and practically
expressed by this initiative of the ANIMP Construction Section that proposes a reference for
spreading and divulging industrial plant culture and best practices. I hope that this way of
working will drive the future challenges of the Association.
Nello Uccelletti
(President of ANIMP)
Advances in plant modularisation / 3

Credits and acknowledgements
The book is based on data, opinions, procedures and designs shared by the members of the
ANIMP-ECI Task force on Modularisation. The experience and knowledge emerged during the
vivid and fruitful discussions within the Task Force in the last two years are the real origin and the
strength of the research.
Under these premises, Chapter 1 and Appendix A has been written by Mauro Mancini and Nicola
Careri; Francesco Di Serio and Alistar Gibb significantly contributed to the paragraph on “Discussion
and improvement areas”. A special thanks goes to Paolo Androni who supported the overall
activity within his MS thesis work at DIG – Politecnico di Milano - and to Tristano Sainati (FARB re-
searcher) for the design of the questionnaire.
Chapter 2 has been written by Paolo Cremonini, Gabriele Scandolo and Loris Giovannini from
Fagioli S.p.A.
Chapter 3 and Appendix B has been written by Federico Perotti and Raffaele Ardito, who are in-
debted to Alessandro Palmeri for the enlightening comments and suggestions. The prototype
structure described in the Appendix was made available by Foster Wheeler Italiana Srl with the
essential contribution of ideas and suggestions by Flavio Vitalini. Maria Chiara Padovani and Fran-
cesco Riva supported the development of the case study within the MS thesis work performed at
DICA – Politecnico di Milano.
All the industrial members of the ANIMP-ECI Task Force reviewed the entire text under the coordi-
nation of Marco Spinelli.
ANIMP and Politecnico di Milano (under the support of the FARB project on modularisation)
financed the activity including the graphic setting and the printing of the volume and involved
Companies supplied picture and reported documents.
Mauro Mancini
(ANIMP Construction Section Coordinator)
4 / Advances in plant modularisation

Index
Introduction ............................................................................................................................................6
Chapter 1 : Survey on modularisation management .............................................................9
1.1 Literature overview ...........................................................................................................................9
1.2 Survey features ...............................................................................................................................11
1.3 Results............................................................................................................................................12
1.3.1 Definition, application areas and pro&con stakeholders...............................................................12
1.3.2 Driving factors and objectives.......................................................................................................14
1.3.3 Constraints...................................................................................................................................19
1.4 Discussion and improvement areas.................................................................................................23
Chapter 2: module handling and transportation ...................................................................29
2.1 Module handling............................................................................................................................29
2.2 Marine transportation criteria.........................................................................................................29
2.2.1 Sea motion criteria, grillage and seafastening design ..................................................................30
2.2.2 Sea transportation vessels load-out and load-in ..........................................................................34
2.3 Transportation by SPMTs.................................................................................................................38
Chapter 3: Structural design aspects: a case study................................................................43
3.1 Introduction....................................................................................................................................43
3.2 General issues ................................................................................................................................45
3.2.1 Weight reduction: welded joints vs bolted joints..........................................................................45
3.2.2 The determination and treatment of wind actions........................................................................47
3.2.3 Stiffness: horizontal bracing as a prototype problem of code application .....................................47
3.2.4 Stiffness 2: composite steel-concrete columns and the versatility concept....................................48
3.2.5 Seismic design strategy for Ultimate Limit States..........................................................................49
3.2.6 Seismic design strategy for damage limitation..............................................................................50
Conclusions .......................................................................................................................................51
References ........................................................................................................................................54
APPENDIX A: Questionnaire...............................................................................................................55
Interview on modularisation management ............................................................................................55
APPENDIX B: Case study ....................................................................................................................62
B.1 Analysis of the structural response to wind loads ............................................................................62
B.1.1 Wind action: the structural factor .................................................................................................62
B.1.2 Structural factor for the original layout.........................................................................................63
B.1.3 Modified layout: horizontal bracing system..................................................................................64
B.1.4 Further modification: stiffened columns ......................................................................................68
B.1.5 Validation via dynamic analysis ....................................................................................................68
B.2 Seismic analyses .............................................................................................................................71
B.2.1 Proper evaluation of the behavior factor ......................................................................................72
B.2.2 Operational details of seismic analyses........................................................................................77
B.2.3 Safety assessments for the original layout....................................................................................81
B.2.4 Design suggestions for increasing the seismic performance .........................................................84

Introduction
“Is modularisation a suitable execution strategy for our industrial plant project?”
This is the question that is being asked more and more among plant EPC contractors and clients.
Modularisation is a forced choice in the case of off-shore installations and a cutting-edge design
paradigm within the nuclear plant sector [1] [2] [3]. This construction approach in the last few
years has been applied also by petrochemical, chemical, gas processing and oil refining onshore
plants. In 2008, North West Shelf Venture Phase V in Western Australia was the first LNG plant to
realize this conception, but since then several Liquefied Natural Gas (LNG) plants have been
using this execution strategy as well [4]. Other recent examples of plant modularisation are the
expansion of an existing refinery that Saipem is performing in Suriname(1), the new units realized
by Foster Wheeler in a refinery in Belgium as well as three plants designed by Technip in North Al-
berta (Canada) for the Horizon Oil Sand Project between 2005 and 2012(2).
What is a module?
A plant module is a transportable, prefabricated/preassembled steel structure containing
static and rotating equipment, piping, instrument, electrical hardware and associated cabling
that can be constructed and pre-commissioned in areas with controlled conditions that differ
from the final location. These areas are called yards. After the assembly in the yard, the modules
are transported ( by road or sea) at site for final installation and integration in the final plant.

Introduction
These type of projects stimulated ongoing analysis and a still developing literature, focusing on a
modular approach to construction, its basic dimensions and the main drivers that lead to the
choice of this particular execution strategy.
Nevertheless, due to the strict confidentiality that covers such projects, three elements are still
missing from a full understanding of modularisation potential:
» a systematic comparison between concepts expressed in literature and actual practices;
» an assessment of these practices aimed at identifying gaps to fill in order to make modula-
risation a fully operational and ready to use approach;
» an analysis of the identified technical criticalities focused on providing general and on

hand design solutions
In order to fulfil the threefold purpose of this work a survey was first conducted among the main
Italian Engineering, Procurement and Construction (EPC) contractors and some service providers.
This survey allowed the comparison of drivers, barriers and dimensions generally associated with
Figure 1.1 - One of the pre-assembled units for the Woodside-operated Train V Phase
5 LNG expansion facility at Karratha, Western Australia. (Source: Foster Wheeler)
(1) http://www.saipem.com/site/home/press/by-year/articolo6121.html
(2) http://www.technip.com/en/press/technip-awarded-two-major-contracts-horizon-oil-sands-project-canada

Introduction
modularisation with the actual perceptions of a significant sample of the Italian EPC supply chain.
Next, starting by considering some technical criticalities highlighted by the survey, a case study
was developed dealing with some crucial issues in module design, in order to provide valuable
and innovative engineering solutions.
The exposition of the work is organized as follows: chapter one regards the survey and consists of
four sections: the first one presents the literature review, with particular reference to drivers and
criticalities generally associated with modularisation; section two explains the questionnaire
structure highlighting methodological aspects, enlisting the companies involved in the survey and
specifying the interviewed roles; the third section presents and analyses the interview results;
section four discuss the results identifying improvement areas and some design gaps to be filled
in order to enhance modularisation operability and application range. The second chapter is
devoted to the criteria for module handling, including sea transportation and land transportation
via SPMTs. The third chapter describes instead the proposed technical solutions from the structural
point of view, on the basis of the analysis of a case-study. More specifically, the third chapter
treats some specific issues which are of crucial importance in the achievement of the following
objectives, strictly related to modular plants: preservation of the structural safety; optimization of
the structural weight; increase of the structural versatility. The suggested modifications to the
structural layout of the case-study can be considered of general validity for modular plants. Finally
chapter four summarises results and recommends key future areas to develop.
Figure 1.2 - Modularized steam
reformer furnace for hydrogen
production unit, designed and
built by Foster Wheeler for a
plant in Nigeria.
The 750-ton weight heater was
delivered completely assembled.
(Source: Foster Wheeler)

1 | Survey on
modularisation
management
1.1 Literature overview
Recent literature on modular construction agrees that modularisation is everything except that silver
bullet [5]. Caswell et al. [6] reinforces this point eloquently, stating that: modular construction is
an appropriate execution strategy, particularly where the following circumstances apply [5] [7]:
» Very high labour cost at site
» Very low labour productivity at site
» Wide lack of skilled manpower in the region/area of the site
» Lack of adequate infrastructure to host a high number of expat workers
» Restrictions on maximum number of allowed expat workers
» Constraints on maximum number of workers simultaneously operating at site
» High probability of severe weather conditions during the construction phase at site
» Serious safety/security concerns at site
» Need for a project crashing
» Significant need of resources subject to high competition among the company’s project
portfolio

Survey on modularisation management
» Environmental, legal or regulatory constraints at site

Under these circumstances, which come together particularly in remote areas as well as in areas
where there is a high demand of labour, a modular approach may offer significant advantages
compared to a traditional stick-built construction strategy. These advantages are derived mainly
from shifting a considerable amount of work from the site to one or more fabrication yards located
in strategic areas [4] in which sufficient skilled and cost-effective construction manpower is
available [5]. Yard fabrication allows modules to be produced in an efficient, safe and controlled
environment using a lower cost skilled workforce and achieving high quality standards. Furthermore
performing the construction in different locations simultaneously and bringing the modules to
the site afterwards, may yield savings in project execution time. The above advantages have been
summarized in Table 1.1
Modularisation benefits
Schedule Savings Higher Safety
Improved quality Higher Security
Social/Environmental Impacts reduction Lower Manpower costs
Reduction of weather impacts

Table 1.1 - Modularisation benefits generally recognized by literature
Besides these boundaries dependent advantages, modularisation also has some disadvantages
that may affect the whole project lifecycle. A list of the disadvantages most frequently associated
to modularisation in literature, is provided in Table 1.2
Modularisation drawbacks
Higher engineering effort Higher structural costs
Higher transportation costs Higher need for infrastructure
Local content impacts ...
Table 1.2 - Modularisation drawbacks generally recognized by literature
What arises from this literature overview is a clear perception of modularisation as a strategy to
reduce overall project risk, under some bound and determined hypothesis. What is not clear is
how much this common perception is shared by the EPC contractor supply chain, and how much

weight companies and operators give to the above aspects. Furthermore is not clear which gaps
need to be filled in order to completely deploy modularisation’s potential. Research was conducted
to shed light on these matters by means of a survey among some main actors of the Italian EPC
contractors supply chain.
''HǏQLWLRQV
HǏQLWLRQV
$PRGXOHFRQVLVWVRIDYROXPHǏWWHGZLWKVWUXFWXUDOHOHPHQWVǏQLVKHVDQGSURFHVVFRPSRQHQWVWR
$PRGXOHFRQVLVWVRIDYROXPHǏWWHGZLWKVWUXFWXUDOHOHPHQWVǏQLVKHVDQGSURFHVVFRPSRQHQWVWR
EEHHDVO\LQWHJUDWHGLQDSODQWGXHWRžZHOOGHǏQHGſLQWHUIDFHV0RGXOHVFRQWDLQSUHIDEULFDWHG
HHDVO\LQWHJUDWHGLQDSODQWGXHWRžZHOOGHǏQHGſLQWHUIDFHV0RGXOHVFRQWDLQSUHIDEULFDWHG
FFRPSRQHQWVRUSUHDVVHPEOLHVDQGDUHFRQVWUXFWHGDZD\IURPWKHFRQVWUXFWLRQVLWH
RPSRQHQWVRUSUHDVVHPEOLHVDQGDUHFRQVWUXFWHGDZD\IURPWKHFRQVWUXFWLRQVLWH
0RGXOHZHLJKWW\SLFDOO\YDULHVEHWZHHQDQGWRQQHV0RGXOHVL]HLVHVVHQWLDOO\GHWHUPLQHGE\
0 RGXOHZHLJKWW\SLFDOO\YDULHVEHWZHHQDQGWRQQHV0RGXOHVL]HLVHVVHQWLDOO\GHWHUPLQHGE\
IIDEULFDWLRQIDFLOLWLHVWUDQVSRUWDWLRQ ORJLVWLFFRQVWUDLQVRIǐRDGLQJIDFLOLWLHVPRGXOHKDQGOLQJKHDY\
DEULFDWLRQIDFLOLWLHVWUDQVSRUWDWLRQ ORJLVWLFFRQVWUDLQVRIǐRDGLQJIDFLOLWLHVPRGXOHKDQGOLQJKHDY\
OOLIWDQGKRRNXSWHFKQLFDOVROXWLRQVPDLQWHQDQFHDQGRSHUDELOLW\VLWHOLPLWDWLRQVDQGFOLHQW
LIWDQGKRRNXSWHFKQLFDOVROXWLRQVPDLQWHQDQFHDQGRSHUDELOLW\VLWHOLPLWDWLRQVDQGFOLHQW
UHTXLUHPHQWV
UHTXLUHPHQWV
0RGXOH
0RGXOH
0RGXOHVZLWKGLPHQVLRQLQWKHRUGHURI[[PHWHUXSWRWRQWKDWFDQEHWUDQVSRUWHGE\
0RGXOHVZLWKGLPHQVLRQLQWKHRUGHURI[[PHWHUXSWRWRQWKDWFDQEHWUDQVSRUWHGE\
UXFNDUHXVXDOO\FDOOHGžVPDOOPRGXOHVſ
WWUXFNDUHXVXDOO\FDOOHGžVPDOOPRGXOHVſ
0RGXOHVZLWKGLPHQVLRQLQWKHRUGHURI[[PHWHUVDQGZHLJKWXSWRWRQVDUHXVXDOO\FDOOHG
0 RGXOHVZLWKGLPHQVLRQLQWKHRUGHURI[[PHWHUVDQGZHLJKWXSWRWRQVDUHXVXDOO\FDOOHG
ODUJHPRGXOHVſ/DUJHPRGXOHVUHTXLUHVHDWUDQVSRUWDWLRQ
žžODUJHPRGXOHVſ/DUJHPRGXOHVUHTXLUHVHDWUDQVSRUWDWLRQ
33$5
$5 $W\SLFDOVWHHOSLSHUDFNPRGXOH3$5LQVWDOOHGRQFRQFUHWHSOLQWKVRUVOHHSHUVHQDEOLQJSLSLQJFDEOH
$W\SLFDOVWHHOSLSHUDFNPRGXOH3$5LQVWDOOHGRQFRQFUHWHSOLQWKVRUVOHHSHUVHQDEOLQJSLSLQJFDEOH
3UH$VVHPEOHG5DFN
3UH$VVHPEOHG5DFN LLQWHUFRQQHFWLRQDQGGLVWULEXWLRQEHWZHHQDOOWKHSODQWDUHDV
QWHUFRQQHFWLRQDQGGLVWULEXWLRQEHWZHHQDOOWKHSODQWDUHDV
PAU $$VWUXFWXUDOVWHHOER[GHVLJQHGWRLQFOXGHRQHRUPRUHSLHFHVRISURFHVVHTXLSPHQWSLSLQJHOHFWULFDO
VWUXFWXUDOVWHHOER[GHVLJQHGWRLQFOXGHRQHRUPRUHSLHFHVRISURFHVVHTXLSPHQWSLSLQJHOHFWULFDO
3UH$VVHPEOHG8QLW
3UH$VVHPEOHG8QLW DDQGLQVWUXPHQWLWHP
QGLQVWUXPHQWLWHP
$SRSXODUPHWKRGRIGLVWULEXWLQJDQGVWRULQJPDFKLQHU\6LPSO\SXWWKHPDFKLQHU\DWSRLQWRI
$SRSXODUPHWKRGRIGLVWULEXWLQJDQGVWRULQJPDFKLQHU\6LPSO\SXWWKHPDFKLQHU\DWSRLQWRI
6NLG3DFNDJH
6NLG3DFNDJH DQXIDFWXUHLVSHUPHQHQWO\PRXQWHGLQDIUDPHRURQWRUDLOVRUDPHWDOSDOOHW7KHHTXLSPHQWFDQWKHQ
PDQXIDFWXUHLVSHUPHQHQWO\PRXQWHGLQDIUDPHRURQWRUDLOVRUDPHWDOSDOOHW7KHHTXLSPHQWFDQWKHQ
P
EHHDVLO\DQGVHFXUHO\WUDQVSRUWHGDQGXVHGDVDXQLW)RUH[DPSOHDǏUHǏJKWLQJVNLGXQLWPD\DOVR
EHHDVLO\DQGVHFXUHO\WUDQVSRUWHGDQGXVHGDVDXQLW)RUH[DPSOHDǏUHǏJKWLQJVNLGXQLWPD\DOVR
HWHPSRUDULO\SODFHGRQWRDYHKLFOHWRHTXLSLWIRULWVWDVN
EEHWHPSRUDULO\SODFHGRQWRDYHKLFOHWRHTXLSLWIRULWVWDVN
Table 1.3 - Table of common terms
1.2 Survey features
As shown in Table 1.4, the questionnaire was completed by six Italian EPC main contractors and
one service provider. Nineteen different managers from five different departments were interviewed
to get insights from each EPC’s business area.
The information was collected through guided interviews. In each interview a broad introductory
discussion was followed by the survey compilation. The twofold objective of the introductory di-
scussion was to explain the main objectives and motivations of the research and to let the inter-
viewees describe their concrete experience with modular projects. This produced new insights on
modularisation derived from on-the-field experiences. The questionnaire (see appendix A) consists
of three main sections:

» Definition, application areas and pro&con stakeholders
» Driving factors and objectives
» Constraints
The collected information was discussed and reviewed within periodic meetings with the members
of ANIMP’s Construction department. The next section presents a synthesis of the survey results.
Survey summary
COMAPANIES INVOLVED ROLES INTERVIEWED NUMBERS OF INTERVIEWS
Engineering
Procurement
Construction
Tendering
Project management
FOSTER WHEELER
Table 1.4 - Survey summary
1.3 Results
1.3.1 Definition, application areas and management stakeholders
In the first section of the questionnaire interviewees were asked to give their definition of modu-
larisation, in comparison to four definitions taken from the literature. The definitions expressed
by the interviewees, attribute to modularisation the subsequent common features:
» plant decomposition according to both constructability and process logic

» prefabrication of the modules in yards other than the construction site
» a need to satisfy transport and assembly requirements

A new definition has thus been proposed. The definition is supposed to include what was
highlighted by the interviewees and therefore to effectively fit the EPC field. The proposed definition
suggests that plant modularisation is:
“the decomposition of a plant in elements according to construction and process logic. These elements,
or modules, meet transportation and assembly criteria and are fabricated and tested in fabrication
yards that differ from the construction site”
The interviewees were also asked to choose, assigning a score from 0 to 100, possible areas emerged
from literature where modularisation could be applied. The results are shown in Figure 1.3.
AREAS OF INTEREST MEAN

0 63 13 3 0 0 5 0 0
100
50
0
ct
u
n
rod
tio
on
eP
nta
ti
ity
l
iza
gib
on
me
bil
ct
ce
an
cti
an
du
cu
pa
vi
nt
Org
Ser
Fun
Int
Pro
Do
Ca
Pla
Yar
Figure 1.3 - Average weighted modularisation areas of interest
The main identified area of application for modularisation, is the “Plant”; it yielded an average
score of 63/100, due to the critical impact of plant modularisation on the whole project. Beside
this more than obvious result, it is interesting to observe that the second area of interest is the
“Yard”, with an average score of 13/100. This means that some of the interviewees testify the
need for functional reconfigurability of the shipyard’s areas. The need for a high degree of
scalability in production capacity was also stressed. These yard’s features have to be considered
both as organisational principles for property yard and parameters that address the choice during
the yard selection process. Other areas of interest like “Product”, “Intangible Product”, “Capabilities”,
“Service”, “Organization”, “Function” and “Documentation” yielded non-significant scores.

Figure 1.4 - Installation at site of a module fabricated by Saipem at Arbatax yard

for a refinery in Suriname. (Source: Fagioli)
The interviewees were then asked to indicate which internal/external stakeholders usually advocate
for or against a modular approach. The Construction Department is by far the internal stakeholder
more often identified as a modularisation “advocate”. The interviewed sample attributed this fact
mainly to the easier and safer construction activities enabled by modularisation. The internal sta-
keholder usually associated to stances adverse to modularisation, is instead the Engineering De-
partment, due to the increased complexity associated with module design and a lack of confidence
with the approach. Regarding the external stakeholders, the client is the one who usually takes
the final decision whether modularise or not. He could be in favour or against modularisation,
according with his identified priorities and objectives. Some of the interviewees stated that clients
usually link modularisation to higher costs but faster delivery times. Local governments could be
likewise for or against modularisation, mostly depending on the will to maximize the local content
or otherwise minimize the social impact of construction activities at site.
1.3.2 Driving factors and objectives
In order to establish a priority ranking among factors that drive a project towards modularisation,

the interviewees were asked to summarize them through an open-ended question, assigning a 0-
100 weight to each of the mentioned driver. The drivers were then gathered under eleven main
categories: Costs, Schedule, Site Conditions, Labour, Social/political, Safety, Constructability,
Quality, Competitiveness, and Logistics.
DRIVERS
!"#$%%&&'(
48
40 38
28
23 22 21
15 14
7
1
SCHEDULE
SITE CONDITIONS
LABOUR
SAFETY
SOCIAL-POLITICAL IMPACTS
CONSTRUCTABILITY
COSTS
QUALITY
LOGISTICS
RISK REDUCTION
COMPETITIVENESS
Figure 1.5 - Average weighted modularisation drivers
As shown in Figure 1.5 there are three modularisation drivers that yield an average score higher
than 30/100:
» Schedule: with an average score of 48/100, schedule savings are recognized as the main
driver for modularisation, especially when the customer is more focused on fast scheduling in
lieu of cost containment. Most of the interviewees agree that this advantage could be achieved
even if the circumstances presented in the literature review chapter do not occur. Schedule
savings derive from the opportunity to allocate project construction workload to different fabri-
cation yards. Furthermore, some of the interviewees state that modularity allows fully exploitation
of the capabilities of subcontractors that can design and produce finite modules. Further time
savings may be achieved by paralleling design and detail engineering activities. Last but not

least, especially for packages that completely execute a specific function, some commissioning
activities can be done in the yard, with obvious advantages in terms of time and money.
» Site conditions: this category includes weather conditions, site remoteness and security
issues at site and received the second highest average score (40/100). Indeed, severe climate
conditions for extended periods of time may hamper construction, introducing significant
delays and cost increases, especially in locations with strict labour regulations (e.g. Alberta,
Canada). Likewise construction sites located in highly dangerous and risky areas require a
huge effort in site security and workers/asset protection. Site remoteness with its lack of infra-
structure and adequate facilities to host expat workers, is also an important factor. Finally, site
conditions are particularly important since they usually have indirect effects on manpower
availability. Modularisation allows mitigation of the above criticalities, reducing the amount of
work performed at site.
Figure 1.6 - A view of the Horizon Oil Sands Plant, Fort Mc Murray, Northern Alberta,
Canada. In the last decades modularisation has been largely adopted in order to
mitigate the impact of severe weather conditions at site. (Source: Technip)
» Labour: The availability of low cost skilled manpower at site as well as an industrial area
able to adequately support plant construction is a key factor for a successful project. The inter-
viewees assigned an average weighted score of 38/100 to this driver. Remote site locations and

areas with poor industrial development impose massive importation of expat workers, with the
well-known issues in terms of costs and local content. On the other hand, areas characterised
by intensive construction activities (e.g. Houston area) or disproportionate workers’ wages (e.g.
Australia), could make impossible to find available man power at reasonable costs. Modularisation
provides effective solutions concerning these criticalities, allowing work to be shifted to areas
where low cost manpower is largely available and yards or shops productivity rates are much
higher. Interviewees point out that another relevant labour-related benefit of modularisation,
is the reduction of the maximum number of workers simultaneously present at site, with positive
consequences in terms of site congestion, work efficiency, costs and resources utilisation balance
within the company project portfolio.
Figure 1.7 - One of the 160 modules fabricated in Thailand and then transported to
Gladstone, Australia for the Queensland Curtis LNG project. Labor cost and skilled
workforce availability are crucial drivers for the modularisation of the recent
Australian LNG plants. (Source: Fagioli)
The interviewed sample considers the enhanced Safety (average weighted score: 28/100) a me-
dium importance driver for modularisation, since fabrication yards are usually safer and a more
controlled environment than construction sites. Also, Socio-Political factors (average weighted
score: 23/100) were found to be a relevant driver. Even medium size plants may indeed require
thousands of workers simultaneously operating at site. This has serious implications on social
fabric, especially in small countries or regions where the government requires projects to minimise

the impact of construction activities on the local population. Also Constructability-related con-
cerns, like layout constraints at site, yielded a significant average weighted score (22/100).
It is not surprising that “Costs”, are only seventh in the drivers ranking. If none of the circumstances
listed in paragraph 2 occurs, modularisation itself certainly implies higher costs in terms of
structural steel, welding and transport costs. So what has driven our interlocutors to assign a
21/100 an average weighted score to “Cost” is that, in some specific scenarios, modularisation is
the only feasible execution strategy. In an ideal world the stick built approach wins ‘hands down’
in terms of first cost. But in the real world, especially for challenging remote located projects, mo-
dularisation may be the only feasible strategy to reduce overall project costs.
Other drivers like Quality, Logistics and Competitiveness were less stressed by the inter-
viewees.
The interviewees were then asked to assign a weight to sixteen disaggregated objectives usually
related to plant modularisation.
OBJECTIVES
!"#$$%&'()*++,
51 51 47 46 45
40 40 38
35 33 32
27
23
19 19
11
Reducing and shortening of MPW peak
Increasing construction efficiency
Reducing project delivering time
Reducing project cost
Enhancing innovation
Enhancing market competiveness (e.g. better services)
Enhancing project manageability
Reducing complexity
Balance standardization with customization
Reducing design and/or manufacturing efforts

Reducing imported MPW and “staking” of craft
Mitigating social impacts at site due to construction
Reducing risk
Mitigating the lack of lay-down areas constraints
Increasing sustainability of the site works

Increasing safety
Figure 1.8 - Average weighted modularisation objectives
The analysis of the results illustrated in 1.8 confirms that the interviewees assign a primary impor-
tance to disaggregated objectives related to skilled manpower availability and cost, schedule re-

duction and site context issues.

1.3.3 Constraints
As for the drivers, the interviewees were asked an open question to list the major constraints as-
sociated to the adoption of a modular construction approach, assigning to them a score from 0 to
100. What emerged from the results, reported in Figure 1.9, is that there are three most relevant
constraints:
» Module design issues
» Module transport issues
» Issues related to the higher complexity associated to modular projects

In regard to module engineering issues (average weight: 24/100), the need to provide the fa-
brication yards with the engineering deliverables as complete as possible and the opportunity to
CONSTRAINTS
!"#"$"%"&"'"(")*" "*"+",",".".
24 23
21
16
13 13
7
6 5 5 4
2 2 1 1
ENGINEERING ISSUES
TRANSPORTATION
INCREASE IN COMPLEXITY
INFRASTRUCTURES
SITE ACCESIBILITY
PROCUREMENT ISSUES
WEATHER CONDITIONS
LIFTING
REGULATIONS
LOCAL CONTENT
COSTS
SUB-CONTRACTOR KNOW-HOW
YARD AVAILABILITY
TECHNOLOGICAL CONSTRAINTS
FIT-UP
Figure 1.9 - Average weighted modularisation constraints
parallelise the construction activities with fabrication enabled by modularisation, induce an anti-
cipation and a compression of the engineering phases which may not be easily manageable.
Furthermore, the need to accomplish the detail engineering as soon as possible on the basis of
still significantly uncertain information may drive the adoption of a particularly conservative design
approach, with obvious cost increases.

Also, it is possible that the module design itself implies constraints related to the plant complexity
or to the lack of the designers’ familiarity with a modular approach. Adopting this kind of approach
for module design forces the engineering teams, in particular the detail engineering teams, on
one hand to anticipate and reduce the delivery time of its deliverables; on the other hand detail
engineering is also forced to work on the bases of partial information, which are subject to great
uncertainties and feedbacks mainly relative to transportability, lifting and site accessibility analysis
as a partial solution to this issue. Some of the interviewees identified, the opportunity to involve,
from the very first phases of the engineering, a specialised module supplier, in order to lighten
the work burden on the main contractor engineering department and in order to solve problems
related to the main contractor not being familiar with module design.
Next, module transport issues (average weight: 23/100), are considered by the interviewees a
major constraint. They have indeed a direct effect on module design. In particular, modules that
need transportation by sea, are seen to require a design to deal with strong dynamic forces. So
modules are usually provided both with bracings and structural reinforcements. The equipment
disposition is aimed at maximising stability as well. This has consequences in terms of higher
costs related to raw materials and increased design and welding activities.
Figure 1.10 - Technical drawing of the stowage of modules produced in Italy and
then transported by sea to South America. (Source: Fagioli)

Furthermore, carrying modules that can reach even 6000 tonnes requires the availability of
extremely expensive transport systems and means that only a very limited number of carriers are
able to tender. The size and weight of the modules make difficult the identification of viable
transport routes, because of physical constraints, local regulations and the lack of adequate in-
frastructures (average weight: 16/100 ). This explains why a large portion of the respondents
recognised the relevance of the lack of infrastructure. A proper selection of the fabrication yard
should, therefore, carefully consider the means and the route to be used for the transportation of
the modules from shop to site, identifying all the possible physical and regulatory constraints as
well as any action needed to make transport feasible. It is not unusual that transport of the
modules requires the construction or the adaptation of roads and bridges, with high costs even
for a few miles. Interventions of this kind may also be required in order to enable the accessibility
of the site (average weight: 13/100 ), such as the construction of docks suitable to download
modules from the barges and the adaptation of construction site access roads.
The increased complexity related to the adoption of a modular approach (average weight:
21/100) is considered particularly relevant by the interviewees. From the project management
standpoint, a situation where construction activities are divided between those carried out at the
site and those at a fabrication yard , is reflected in duplicated planning and supervisory activities:
the project is therefore characterised by a double schedule (the more yards involved, the higher
supervisory effort required). This increase involves not only the EPC contractor but also the client
company that may be forced to relocate its staff to multiple locations. Particularly critical is the
transport planning activities. Errors in the estimation of module delivery time could lead to the
means of transportation not being able to wait till delivery. This then would require rebooking of
the transport systems which may not be available for several months.
Procurement activities (average weight: 13/100) are significantly affected by the adoption of a
modular approach. This involves a further step in the supply chain, since materials and equipment
must be first delivered to the yard and then transported (within modules) to the site. In addition,
the need for parallelisation of the activities of engineering or even the possible outsourcing of such
activities for specific modules, leads to a reduction in the volume of components bought from the
same supplier, with obvious scale diseconomies. Some respondents also associate the parallelisation
of engineering with a possible negative impact on the degree of commonality and standardisation
of components within the different units of the system. Harmful effects on the costs of the components
can also be caused by the aforementioned need to accelerate detailed engineering at the expense
of accuracy in the purchase specifications and the possibility of a careful selection of suppliers.
Finally, a factor of considerable complexity is the management of the module installation sequence.
In fact, installation sequences that are not robust and flexible with respect to unexpected circumstances
may produce situations in which the failure to complete a particular critical module causes the in-
terruption of the construction activities. This could have severe repercussions on the delivery time
of the project.

Figure 1.11 - Module installation operation at two different sites in Canada

(Source: Fagioli/Technip)

1.4 Discussion and improvement areas
One of the main results of the survey was to highlight the main gaps to be filled from a managerial
and technical point of view in order to deploy the full potential of modularisation.
Analysing the survey results at an aggregate level modularisation emerges as an execution strategy
that produces impacts of capital importance on each phase of the project.
Moreover, although the modular approach is already widely known by the main players in the
EPC sector, a comprehensive and shared understanding of it within companies is still missing.
The survey clearly highlighted some trends in the identification and prioritisation of constraints
and drivers associated with modularisation, but the importance associated by respondents to
these factors vary significantly. This is confirmed by the fact that the driver of greatest significance
has an average weight of just 51/100. The most relevant criticality however gained an average
weight equal to 24/100. This effect can also be read as a substantial dependence of the interviewee’s
Figure 1.12 - A modularised

cold box for an LNG plant at
Hammerfest, Norway . The
unit, that had a weight of
2700 tons and was 60 meters
high, was transported from
Belgium to Norway by barge
and then a semi-submersible
ship. (Source: Fagioli)

answer by the specific projects he/she had been involved in and obviously by the overall slight
common view shared in the company.
From the EPC contractors’ standpoint, plant modularity appears often to be a forced condition
(maybe due to an explicit request of the client or by the boundary conditions of the project) more
than a real strategic and operational decision. In this framework the Italian EPC sector needs to
encourage the development and the diffusion of a modularisation culture, but at the same time
to adopt managerial approaches that minimise the possibility of neglecting the fundamental prin-
ciples of modularisation. The construction department is probably the most suitable stakeholder
to promote and drive the consolidation of such a corporate culture.
According to this study, new project management approaches dedicated to modular projects have
to be characterised by:
» Decision making tools that, from the earliest project phases, allow estimating, the suitability
of a modular approach with regards to the fundamental dimensions of the project. Many of the
interviewees declared that, making the decision whether to modularise or not “sooner rather
than later”, is crucial. For this reason the first constraint of such a decision making tool is the
uncertainty of the data received as input. Perhaps the scoring model developed by CII (Con-
struction Industry Institute(3)), may be considered an embryonic attempt to develop tools of
this nature, but updates and improvements of this tool are hugely desirable, and what Saipem
[4] developed for this purpose is definitively a step forward in this direction.
» New approaches to engineering activities that allow to exceed the vision of modularity
just as a "partition" and splitting of a standard plant. Engineering activities customised for mo-
dularisation should indeed foster as much as possible the anticipation of modules’ interfaces
definition; both from a structural and functional standpoint. Feed-back loops existing between
modules’ design and transportability, lifting and site accessibility analysis have to be evaluated
from the very first engineering phases. Process managers have to be involved to explore new
technologies for module design enhancing functional completeness of each module. This could
boost the commissioning activities at the fabrication yard, ideally resulting in ready for start-up
modules. Also adopting golden weld approaches and precast foundations may result in minimized
hook up activities.
» Procurement approaches able to not underestimate the increased complexity induced

by modularisation, expressing a particular effort in the coordination with the Engineering de-
partment. Procurement and engineering departments should strongly cooperate in order to
minimise the reduction of components commonality induced by modularisation. The definition
of equipment and homologous components specifications that follows milestones shared by
all the plant modules is also desirable. This would make possible to limit the negative effects
on the amount of purchased components. Mitigating these effects is particularly complex when
the the design and the manufacturing of parts or of all the modules are subcontracted. In this

Figure 1.13 - 3D study of the installation of a 4800 tons module for a regasification
terminal at Rovigo, Italy. (Source: Fagioli)
case a valuable strategy is following a centralised procurement as far as possible or at least

strongly coordinated. The interviewees have different feelings regarding the best operational
strategy for carrying out fabrication yards’ Procurement. On the one hand it is argued that for
simplicity reasons, construction activities should begin only once all the necessary materials
and equipment for the modules production has been delivered. On the other hand someone
prefers an approach where the module is assembled as soon as what is necessary to the fabri-
cation of the structural elements becomes available.
A common point among the interviewed was the need to develop updated and comprehensive
short lists of vendors and subcontractors holding the needed capabilities to provide both
process packages and entire modules. Often the suppliers specialised in the fabrication of the
main components or process unit within a skid/module, have no adequate capabilities to
execute the entire module. Engineering and Construction departments should effectively support
the procurement department in such mapping, especially during the evaluation of fabrication
yards and not simply components vendors/suppliers.
(3) http://construction-institute.org/

» Improved sequence of installation analysis. As mentioned the delay of even a single

critical module in the sequence of installation, may have such severe impacts to nullify the
potential benefits of modularisation in terms of reduction of the delivery times. Flexible and
robust installation sequences are crucial for a modular plant project, and arguably this factor
should not be overlooked even in the plant layout design.
» Enhanced project management and supervision tools. The survey highlighted

some perceived inadequacy of tools currently in use in addressing modular plant projects.
Some project controllers highlighted the need to rethink the WBS and OBS templates generally
used in the company, in order to distinguish two or more sets of activities: those performed at
site and those performed at the fabrication yards.
» New organisational structures: the organizational structure of a modularised plant

project has been deemed a critical success factor by the interviewee as it refers to complex
work that is divided in various parts as first and then recombined. Automotive and , computer
sectors have been on the cutting edge of modular studies, but the main principle can be
applied to any type of business, independently from the dimension (large or small sectors).
The first basic tenet of the modularisation organization structure is that each stakeholder of the
project is part of the business as a whole. Like modular furniture, each piece has a place and
distinct purpose, but aimed at the same objective.
Figure 1.14 - Panoramic view of a modularized refinery performed in Suriname. An

accurate analysis of the installation’s sequence of the modules is a key success
factor for a modularized plant’s project. (Source: Saipem)

» Modularisation projects are executed basically through three main type of stakeholders,
i.e. the main contractor project management at the Home Office, the fabrication yard Contractors
and the site erection contractor, that is the receiver of the produced modules. The equipment
and Materials Suppliers are common to all Parties. To be effective, the Organization should
foresee that each party is to stand strong its own, so that it can better support the business as a
whole and execute the work together seamlessly. It means that organization and resources of
the EPC Project Management Team should be self-governing at each modularisation yard, i.e. it
it should replicate the organization of the main project management team and have the
resources and operating functions for managing its portion of the project and for dealing directly
with all parties involved in its own specific module, including material suppliers and home
office Engineering team. The above organization structure has been recommended by almost
all the interviewed construction people.
Site
Erection
H.O. Eng. Contractor H.O. Eng.
& Design & Design
for Yard 1 for Yard 2
Yard 1 EPC Mgt.

Site EPC Yard 2 EPC Mgt.
Mgt.
Fabrication
Contractor
Yard 1 Equipment
& Materials Fabrication
Suppliers Contractor
Yard 2
»
Figure 1.15 - Modularized organizational structure
Proper contractual forms. As it is known, the implied covenant of good faith and fair
dealing is a general assumption that the parties of a contract will deal with each other honestly,
fairly, and in good faith, so as to not destroy the right of the other party or parties to receive the
benefits of the contract. This is implied in every contract in order to reinforce the express
covenants or promises of the contract. Nevertheless, in some cases, conflicts still occur. When
this occurs, a negative impact is suffered by the project, that in some cases could lead to di-
sruptions and delays.

For reducing or hopefully avoiding the risk of the above conflicts, the contract should be fair but
also and mainly it should fit the project configuration. Several type of Contract for implementing
projects exist that, for sake of brevity, can be classified under two main categories, i.e. Lump Sum
Turn Key (LSTK) and Reimbursable Contract (RC). The decision is the responsibility of the customer
who, in taking the decision, should consider the significance of the area of uncertainties for the
definition of the economics and time schedule, aiming to fairly share the risk of the project imple-
mentation. LSTK Contract could appear simpler to be managed by the customer, but it presents a
higher risk of conflict in a modularisation project compared with a stick built one. This is basically
because of its rigidity which does not fit the complexity of the modularized project. It has been ex-
perienced that Reimbursable Contract fit the modularisation project configuration better, given its
flexibility as well as its sharing of risk between customer and main contractor.
The above considerations can be extended also to the fabrication contractor who are normally
managed by the main contractor as a subcontract. In the case of the modularized project, the
relative weight and the incidence of the fabrication contractor on the entire project does not
make the subcontract the perfect tool for managing the work. A certain type of partnership, such
as Joint Venture or Consortium, seems more suitable. Its definition should be consolidated since
the time of the submission of the bid to the customer and should contain the share of the costs
and schedule, so that the mutual interest of all the parties reduce the risk of conflicts.
Modularized project will be concluded successfully if a supply chain is built up between the main
contractor, the fabrication contractor and the erection contractor.
Finally, the survey returned relevant output about possible tactical (short term) and strategic (long
term) actions to support the optimal use of the modular approach. Some examples of this could be:
» The development of partnership and alliances with skilled manufacturers, involving them
not only in the manufacturing but even in the module design. An important expected outcome
of these partnerships is: a lighter workload on the main contractor’s engineering department
resources. Negative effects on procurement related to this increase in the subcontractors’ con-
tribution should be prevented, focusing on new and enhanced coordination tools (e.g. system
engineering approches).
» The evaluation of investment opportunities in the fabrication yard site at strategic locations
As mentioned in 1.3.3, the survey has shown how the engineering design is one of critical factor for
the implementation of a modular approach, because of not yet consolidated familiarity of the
designers with this plant configuration. A further critical factor is represented by the transport and
lifting impacts on structural aspects (and therefore on the project cost). This work identified and
prioritized the main dimensions of modularity and outlined some areas of development for the full
exploitation of the approach. In this light, in the next chapter problems and criteria related to module
handling and transportation are reviewed. Furthermore, the study aimed to provide a technical con-
tribution to overcoming some critical issues related to modules design as covered in the third chapter.

2 | Module
handling and
transportation
2.1 Module handling
Normally modules are land transported by Self Propelled Modular Trailers (SPMT) and loaded on
to and unloaded from sea transport vessels; this can be done by lifting, using cranes aboard
geared heavy lift ships, or by the Roll on – Roll off (Ro-Ro) method using SPMTs aboard barges or
Ro-Ro ships. Tandem lifts may be possibly subject to structural design and/or use of spreader
beams (special lifting devices between hook and module, i.e. beams to maintain lifting slings
vertical and separated, to generate only vertical lifting forces in the module).
When very heavy, a module can be loaded / unloaded using skidding systems (skid ways + strand
jacks or skid shoes for a more controlled operation).
All module transport, loading, sea transport and unloading operations are subject to review and
approval of a Marine Warranty Surveyor (MWS), appointed by Client. The MWS’s intervention is
not required in the case of overland transportation.
Sea transportation plays a leading role in module delivery; in addition it needs special attention
for the impact on the module technical characteristics. For these reasons the next paragraph will
be focused on marine transportation criteria and operations.
2.2 Marine transportation criteria
The sea transportation of the modules produces critical cases regarding structural design, and

Module handling and transportation
thus requires detailed modelling.

At first, the adoption, to the possible extent, of standard module support arrangements is highly
recommended, to simplify the grillage arrangements at all stages: in the fabrication yard, on the
sea transport, in the storage area at site and onto foundations.
Temporary support arrangements for sea transport (to be removed when installed onto their per-
manent foundations) should be minimized, while foundations should be designed to enable mo-
dules to be directly delivered and placed upon foundations by SPMTs, possibly with no need for
intermediate jacking arrangements.
The selection of sea transport vessel (ships / barges), the design of grillage, sea-fastenings and
temporary support, as well as the integrity of the module structures themselves must comply with
the Project Design Criteria and are usually subject to review and approval by an independent
MWS appointed by Client.
2.2.1 Sea motion criteria, grillage and seafastening design
The cargo, the internal reinforcement of the cargo, the sea fastenings, the grillage and the vessel
must be designed to withstand the motions and forces resulting from the design transportation
conditions.
Design motions may be derived by means of dynamic response analyses or from model testing.
In all cases, a realistic combination of environmental loads and wave directions, representing
bow, quartering and beam sea conditions should be used. If neither motion simulations nor
model tests are performed, in case of standard configurations and under satisfactory marine pro-
cedures, the default motion criteria may be acceptable.
The Guidelines by the Warranty Surveyors (DNV, Noble Denton, RINA, IMO, etc...), provide the
default motion criteria with reference to the sea transport route and to the design sea state.
As a reference, an example of motion criteria for the definition of the Ships / Barges harmonic
motion, are listed below (ref. GL Noble Denton - ND/0030 – Guideline for Marine Transporta-
tion):
» Roll amplitude 20°;
» Pitch amplitude 10°;
» Vertical acceleration 0.2g;

The standard criteria ND/0030 shown above should be applied in accordance with the following:
» The roll and pitch values listed above should be assumed to apply for a 10 seconds full

Fully cycle Singleamplitude

NATURE OF TRANSPORTATION LOA B [1] L/B [1] Block period Heave
CASE
(m) (m) Coeff (secs) Roll Pitch
1 > 140 and > 30 n/a < 0.9 10 20° 10° 0.2 g
2 > 76 and > 23 n/a any 10 20° 12.5° 0.2 g
Unrestricted 3 < 0.9 30°
(these values to beused unless any of the following apply) ≤ 76 and ≤ 23 ≥ 2.5 10 15° 0.2 g
4 ≥ 0.9 25°
5 < 0.9 30° 30°
≤ 76 and ≤ 23 < 2.5 10 0.2 g
6 ≥ 0.9 25° 25°
7 any ≥ 2.5 any 10 10° 5° 0.1 g

Weather restricted operation in non-benign areas
for a duration <24 hours (see Section 7.9.2 d).
For L/B < 1.4 use unrestricted case
8 any < 2.5, ≥ 1.4 any 10 10° 10° 0.1 g
9 any ≥ 2.5 any 10 5° 2.5° 0.1 g

Weather restricted operation in non-benign areas[2]
(see Section 7.9.2 e). For L/B < 1.4 use unrestricted case
10 any < 2.5, ≥ 1.4 any 10 5° 5° 0.1 g
Inland and sheltered water transportations Equivalent to 0.1 g
(see Section 7.9.2 f). For L/B < 1.4 use unrestricted case 11 any < 2.5, ≥ 1.4 any Static in both directions 0.0
Independent leg jack-ups, ocean tow on own hull.

For L/B ≥ 1.4 use unrestricted Cases 1 to 6 12 n/a > 23 < 1.4 n/a 10 20° 20° 0.0
Independent leg jack-ups, 24 hours or location move.

For L/B ≥ 1.4 use unrestricted Cases 7 to 8 as applicable 13 n/a > 23 < 1.4 n/a 10 10° 10° 0.0
Mat-type jack-ups, ocean tow on own hull.

14 n/a > 23 < 1.4 n/a 13 16° 16° 0.0
For L/B ≥ 2.4 the pitch angle may be reduced to 8°
Mat-type jack-ups, 24 hours or location move 15 n/a > 23 n/a n/a 13 8° 8° 0.0
Table 2.1 - Default Motion Criteria - ref. GL Noble Denton - ND/0030 – Guideline for
Marine Transportation
cycle period of motion;
» The roll and pitch axes should be assumed to pass through the center of floatation of the
sea transport ship
» Phasing shall be assumed to combine, as separate load cases, the most severe combinations of:
- roll ± heave
- pitch ± heave
Alternative default motion criteria may be acceptable as provided, for example, in DNV Rules for
the Classification of Ships, January 2003, Part 3, Chapter 1, Section 4, Ref. [12], or IMO Code of
Safe Practice for Cargo Stowage and Securing, 2003 Edition, Section 7, Ref. [16].
For the Grillage, Seafastening and Cargo Design, the load components during transportation to
be considered when analyzing the total forces acting on the cargo, the vessel and grillage and sea
fastenings are those due to:
» The static weight of the cargo;

» The dynamic loads which result from the vessel rigid-body motion in all six degrees of freedom;
» The static component of weight which acts parallel to the vessel deck when the ship rolls or pitches;
» Wind load;
» Ballast distribution in the vessel.

Regarding the loads due to the motions above, the combination of motions that gives the highest
loading in any direction must be considered. If more detailed information is not available (such
as a dynamic analysis taking account of phase relationships to compute acceleration vectors), the
highest loadings resulting from the following motions is combined as two separate load cases:
» Roll, heave and sway
» Pitch, heave and surge

Loads may normally be calculated using the assumption that all motions can be approximated by
sinusoidal functions.
Alternative method is provided by RINA (Rules for checking the arrangement intended for sea
transportation of Special Cargo); the Guideline provides the equations to calculate the accelerations
ax,ay,az in the generic point x,y,z and the relevant loads Fx,Fy,Fz, under the known hydrostatic data
of the ship / barge. In this case the loads already include the static component of weight which
acts parallel to the vessel deck when the ship rolls or pitches.
Structural loading due to green water impact should not be assessed and it will be assumed that
ship selection and/or direction of stowage on the modules on the ships decks will avoid cargo
overhang, and thus the possibility of cargo immersion.
The grillage and sea fastenings has to be designed in accordance with a recognized standard or
code of practice. Wherever possible, the design should be carried out based on the requirements
of one code only.
The sea fastening shall be designed in order that the static stresses in all members do not exceed
the allowable stresses in accordance with AISC (American Institute of Steel Construction) Manual
or other acceptable code. In some case it is allowed to exceed the standard allowable stress
according to applicable recognized International Standards.
The grillage design and layout should take account of any limitations imposed by the load-out
method, for example the set-down height and width of the SPMTs.
The design of the grillage must be based on the loads derived from the vessel motions as defined
above. The relative stiffness of the ship’s frames and bulkheads shall be taken into account. The
effects of superposition of loads shall be accommodated in the design when welds/connections are

Figure 2.1 - Load-in operation of modules at port from RoRo vessel and transfer on
barge to reach final location at site one by one. (Source: Saipem)
made between the grillage and ship’s deck following load-out.

The purpose of the sea-fastenings is to secure the cargo during the voyage so that neither the cargo
nor ship suffers damage as a result of the loadings derived from the ship motions caused by the
design environment conditions. Primary sea-fastenings shall be designed to be removed easily
without damage to the cargo. During and following the removal of primary sea fastenings, adequate
residual sea-fastening shall remain to safely restrain the cargo until its removal from the ship.
Relevant to Cargo Strength, modules need to have adequate structural strength to be transported
without damage from the maximum loadings resulting from the unit’s motions under the design
environment. Modules shall be generally analyzed as a three-dimensional elastic space frame, in-
cluding appropriate constraints to represent the grillage and sea-fastening support points. The
structural model shall include all primary and secondary members and may take account of the
shear stiffness of floor decking, if appropriate.
In addition to this global analysis, local analysis may also be required with a twofold purpose: to
quantify load effects in localized highly loaded locations (e.g. grillage support or sea-fastening
connection points) and to confirm the adequacy of equipment support frames and saddles and
the connection of such items to the primary module members.The module fabricator operating at
the module assembly site should provide and install sufficient wood covers and plastic wrap /
tarpaulins as and when required, to ensure protection of the module and its components against
the severity of sea transportation conditions, in accordance with the project Preservation Procedures
(based upon manufacturers’ requirements for equipment)

2.2.2 Sea transportation vessels load-out and load-in
Self-propelled vessels can be planned for the transportation of plant modules: they can be either
geared heavy lift ships (capable of self loading and discharge), or flat deck open stern type
module carriers, capable of handling modules across their sterns.
Alternatively barges with tugs can be used. In case of barges or RoRo vessels, loading and unloading
ops by SPMTs or by skidding system should be used.
The ship (or barge) shall be classed by a recognized IACS (International Association of Classification
Societies) Member. The loads induced during loadout, including longitudinal bending, loads on
internal structure and local loads, shall be checked to be within the approved design capabilities.
Mooring attachments and all attachments for jacking or winching shall be demonstrated to be
adequate for the loads anticipated during or after load-out.
Ship stability should be shown to be adequate throughout the load-out operation. Particular at-
tention should be paid to:
» A load-out onto a ship with a small metacentric height, where an offset centre of gravity
may induce a heel or trim as the structure transfer is completed – i.e. when any transverse mo-
ment ceases to be restrained by the shore skidways or trailers.
» A load-out where there is a significant friction force between the barge and the quay wall,
contributed to by the reaction from the pull on system and the moorings. The friction may
cause ‘hang-up’ by resisting the heel or trim, until the pull-on reaction is released, or the friction
force is overcome, whereupon a sudden change of heel or trim may result.
» Cases where a change of wind velocity may cause a significant change of heel or trim during
the operation.
After the module is fully on the ship, then stability should comply with the MWS’s requirements
for marine transportations, and those of contractual technical specifications and ship’s owner. As
a general rule, the minimum ship freeboard during load-out should be 0.5 m plus 50% of the
maximum wave height expected during the load-out operation.
The bundling of openings in the ship’s deck shall also be considered for low freeboards.
The strength of the load transfer bridges or ramps should be demonstrated. Ramps shall be
checked for loads induced by ship moorings and movements and load transfer forces induced by
SPMTs or skidding system.

Figure 2.2 - Technical drawings of the loading area in a fabrication yard. (Source: Fagioli)

Tolerances on ramp movement should be evaluated to be suitable for anticipated movements of

the ship during the operation. Where a ship, due to tidal limitations, has to be turned within the
loadout tidal window, the design of the ramps should be such that when the loaded unit is in its
final position they are not trapped, i.e. they are free for removal. Suitable lateral guides have to
be provided along the full length of ramps.
Sufficient articulation or flexibility of SPMTs should be provided to compensate for level and slope
changes when crossing from shore to ship and vice versa. Calculations shall show that the load is
fully carried by the SPMTs as in the design case, without overstressing the module structure,
especially if the load transfer is between two floating vessels, such as between a ship and an in-
termediate bridge barge arrangement.
The line and level of the ramp and SPMTs shall be documented by dimensional control surveys and
reports as necessary for load control. The line and level have to be within the tolerances defined for
the loadout operation and design. For floating loadouts care shall be taken to ensure that minimum
friction exists between the ship and quay face. Where the quay has a rendered face, steel plates
shall be installed together with the ship fendering system. The interface between the ship and ship
fendering shall be liberally lubricated with grease or other substitute which complies with local en-
vironmental rules.
A loadout is normally considered to be a weather restricted operation. Limiting weather conditions
for the loadout operation shall be defined, taking into account:
» the forecast reliability for the area
» the duration of the operation including a suitable contingency period
» the exposure of the site
» the time required for any operations before or after the loadout operation including ship
movements and moorings, ballasting, system testing, final positioning and initial seafastening
» currents during and following the operation, including blockage effects if applicable
» the wind area of the cargo and the vessel.

Marine Warranty Surveyors (MWS) typically define load-out and offloading operations in classes
according to the tidal conditions. Requirements for design, reserves and redundancy of mechanical
systems will vary according to the class of load-out.
According to Noble and Denton (ref. GL Noble Denton 0013/ND Rev 7 - 22 June 2013 - Guidelines
for Load-Outs) the Class Tidal limitations are the following:
1. The tidal range is such that regardless of the pumping capacity provided, it is not possible to
maintain the ship level with the quay throughout the full tidal cycle, and the loadout must be

completed within a defined tidal window, generally on a rising tide.

2. The tidal range is such that whilst significant pumping capacity is required, it is possible to
maintain the ship level with the quay during the full spring tidal cycle, and for at least 24 hours
thereafter.
3. Tidal range is negligible or zero, and there are no tidal constraints on loadout. Pumping is
required only to compensate for weight changes as the loadout proceeds.
4. Grounded loadout, with tidal range requiring pumping to maintain
ground reaction and/or ship loading within acceptable limits.
5. Grounded loadout requiring no pumping to maintain ground reaction and/or ship loading
within acceptable limits.
Modules shall be designed taking into account static and dynamic loads, support conditions, en-
vironmental loads and loads due to misalignment of the sea transport vessel and quay or uneven
ballasting. For SPMT loading and offloading, the reactions imposed by the trailer configuration
shall be considered. For lifted load-outs, the structure, including the pad-eyes, shall be analyzed
for the loads and reactions imposed during the lift.
The load-out of the quay, quay approaches, wall and foundations have to be demonstrated to the
MWS as being adequate for the loads to be transferred. This can be in the form of historical data
for loading quays. The Marine Offloading Facility (MOF) shall be designed for handling heavy
loads by SPMTs or skidding.
A statement shall be submitted showing the capacity of all mooring bollards, winches and other
attachments to be used for the load-out.
Compatibility between quay strength and elasticity, and the support conditions used for analysis
of the structure, shall be demonstrated as appropriate.
Bathymetric information for the area covered or crossed by the barge during load-out, post-load-
out operations and sail away shall be supplied. Under keel clearance shall not normally be less
than 1.0 m during the period for which the ship is in load-out position. This may be relaxed to 0.5
m, subject to confidence in the lowest predicted water levels, and provided that a check of the
load-out area has been made by bar sweep, divers’ inspection or side-scan survey; these investi-
gations should be sufficiently recent to represent actual conditions at the time of load-out.
Where there is a risk of debris reducing under keel clearance, a sweep shall be made immediately
prior to the ship berthing to ensure that no debris exists that could damage the barge keel plating.
The results of the sweep shall be confirmed by further soundings check around the ship perimeter
after ship berthing. For tidal load-outs, an easily readable tide gauge shall be provided adjacent
to the load-out quay in such a location that it will not be obscured during any stage of the load-
out operation. Where the tide level is critical, the correct datum should be established.

In additional to MWS approval, port or other competent authority approval for the operation
should be obtained, and the required control of marine traffic instituted.
2.3 Transportation by SPMTs
Self-Propelled Modular Trailer (SPMT) are multi-axle trailers designed for the transportation of
large and heavy cargoes. SPMTs are designed in modular construction and can be coupled side to
side and end to end, or remote units can be operated as a single trailer via radio controls. A SPMT
consists of a very strong and rigid chassis, which also acts as a load-carrying platform to which
wheel bogies are attached in pairs, to form the required length of transport. Each wheel bogie
consists of two rubber-tyre wheels and is rigidly fixed to the chassis by a hinged elbow joint,
which is supported on hydraulic rams. This hydraulic ram acts as the suspension for the SPMT
and also provides the lifting capability. Attached to the end of the SPMT is a Diesel driven power
pack, which provides hydraulic power to the various functions of the SPMT.
The SPMT is propelled by hydraulic drive motors, which are mounted on its axles. Hydraulic
power is supplied to each of the drive motors by a pump on the power pack and speed is
controlled via a remote hand operated portable console. Forward and reverse travel is achieved
by reversing the flow of hydraulic oil to the drive motors. Speed of the driven axle is controlled by
Figure 2.3 - The 25 meters deck of an offshore plant on a Self-Propelled Modular

Trailer. (Source: Fagioli)

flow regulators, which prevent the occurrence of over-speed of the wheels. Normal operating
speed of the SPMTs is 5 km/h, but may vary depending on the load and configuration.
The steering of the SPMTs offers major advantages over other forms of conventional hydraulic
trailers. The steering is controlled by electro-hydraulic motors rather than by a series of mechanical
steering rods. Each of the axles of the transporter moves independently and is monitored and
controlled by the computerized control system. This offers total flexibility of steering options with
each wheel bogie able to swing through 260° ( 130°). The position of each wheel is controlled
electronically through the remote operations console and nine steering programs are available.
The electronic steering control can be used for a group of two or more SPMT's to ensure that all
wheel bogies of all the transporters act together and turn about a single point.
Systematic and controlled pumping of hydraulic fluid into or out of each suspension can raise or
lower the transporter bed +300 mm from the normal running height of 1500 mm
Diagram 2.1 - A hydraulic cylinder supports each of the wheel bogies. The cylinders
on each bogie can be linked hydraulically to the other wheel bogie cylinders to form
groups. (Source: Fagioli)
Diagram 2.2 - This allows free flow of hydraulic fluid between each hydraulic ram
allowing the SPMT to negotiate uneven ground, cambers and gradients whilst
maintaining equal loading in each hydraulic ram within that group. (Source: Fagioli)

Diagram 2.3 - Linking all hydraulic rams on a single trailer would result in an
unstable transporter bed with no control. The SPMT is normally split hydraulically
into three separate hydraulic groups (leading to a three-point suspension) that can
be controlled individually from the main power unit. (Source: Fagioli)
Diagram 2.4 - As the SPMT’s negotiates uneven ground, cambers or gradients, the
hydraulic fluid will free flow within each suspension to maintain equal loading
within each suspension group. (Source: Fagioli)
Diagram 2.5 - Additional hydraulic fluid can be can be pumped from the Power Unit
(PPU) into or out of each cylinder group to adjust the level of the SPMT. (Source: Fagioli)

Control of the elevation of the transporter is through the hand operated remote console. Safety
valves protect the hydraulic circuit so that the transporter platform does not collapse in the unlikely
event of hydraulic failure.
Modules are land transported by SPMTs and loaded on to and unloaded from sea transport
vessels by the Ro-Ro method using SPMTs. The load-out path shall be freshly graded prior to
load- out, pot holes filled and compacted, debris removed and obstructions to the load-out path
identified and removed. Where a structure cannot be loaded out directly onto a barge or vessel
without turning, turning radii shall be maximized where possible. For small turning radii, lateral
supports and restraints shall be installed between the trailer and the structure, load-out frame
and cribbage. It is possible (and is often the case) that a site move may be part of the load-out
operation.
Maximum axle loading shall be shown to be within the trailer manufacturer's recommended
limits. ‘Footprint’ pressure on the quayside, linking ramp and ship’s deck shall be shown to be
within the allowable values. Shear force and bending moment curves shall be prepared for the
trailer spine structure, and maximum values shall be shown to be within the manufacturer's allo-
wable figures. Linking ramp capacity shall be demonstrated by calculation and these calculations
shall form part of the load-out procedure.
In general, hydraulic systems should be linked or balanced as a three point hydraulically linked
system to provide a statically determinate support system, thus minimizing torsion on the structure.
In all cases the arrangement shall be compatible to the support assumptions considered for the
structural analysis of the structure being loaded out. A contingency plan shall be presented to
cover potential hydraulic leakage or power pack failure. Stability of the hydraulic system to resist
overturning shall be shown to be adequate, particularly when a 3-point hydraulic linkage system
is proposed. The centre of action of the structure (Center of Gravity COG) shall remain within the
middle quarter of the trailer support base, taking into account any uncertainty in:
1. the horizontal and vertical centre of gravity, with the adequate contingency factors and the
COG envelope;
2. the design wind speed and relevant design wind load;
3. any inclination of the structure/trailer assembly on shore (slope and operational out of ver-
ticality);
4. the predicted inclination of the barge under the design wind and under the ballast operation
(load out/in cases);
5. SPMT’s acceleration and emergency braking for an emergency stop;
6. possible change of heel or trim due to the ballast operation during Ro-Ro phase and due to the
release of hang-up between the barge and the quay, and any free surface liquids within the structure.

Whilst a 3-point linkage system results in a determinate support system, a 3-point support system
is generally less stable than a 4-point support system. Stability for both 3 point and 4 point
support systems shall be documented.

3 | Structural
design aspects:
a case study
3.1 Introduction
As illustrated in previous chapters, there are several structural design aspects which can affect the
feasibility and affordability of modularisation. Within a simple cost-benefit framework, they can
be summarized as follows.
“Direct” costs
» Larger structural cost (material/weight, detailing, etc) due to additional loading conditions
(transportation, lifting, etc)
» Transportation costs
» Need for larger installation means (cranes, etc)

“Indirect” costs
» More complex structural design
» Need to complete structural design in a shorter time
» Need of early interface with transportation/lifting contractor

“Direct” benefits
» Reduction in the on site construction cost
» Reduction of risks associated to onsite construction

Structural design aspects: a case study
» Reduction of project delivery time

“Indirect” benefits
» Better performance (e.g. in terms of stiffness) of the modularized structure
» Better durability (reducing maintenance)
» Better flexibility with respect to overall equipment life cycle (future XXXX revumpings and re-
newals)
In order to investigate such aspects, a typical pre-assembly (pipe rack) taken from a real application
has been studied. Several aspects of the design have been considered and some possible impro-
vements have been proposed. Within the above cost-benefit framework, the following three main
targets have been identified.
1. Reduce weight
To achieve this goal, two main areas of intervention can be explored:

(a) classical structural optimization, which can be obtained both by varying the structural
layout and by working on structural element sizes while preserving the layout. The first option can
easily conflict with the equipment layout and for this reason has been disregarded here. The
second has been pursued, even though code constraints of existing code provisions often hinder
this type of optimization process.
(b) reduction of loads, which can be obtained either by adopting more sophisticated analysis
procedures or by adopting design solutions which are rewarded by the code with a more favorable
load level, the latter being typically the case of seismic loading; both options have been investigated
in this study.
In particular, the working group has extensively studied the theme of a more realistic representation
of various loading conditions for modular structures. However, reduction of loads coming from
the equipment (weight, operation and thermal effects) has not been attempted, even though
some considerations have been made on the action due to PSVs (Pressure Safety Valves); further-
more, activities regarding transportation loads has just begun, with special reference to standard
barges operation, so that no result are presented in this report .
2. Introduce standardization/versatility
Standardization is the key for addressing the need for a more complex design to be performed in
a shorter time (see the above “indirect costs” list); in this light, standardization can be related
either to the actual structure or to the design process itself. It can be argued that it is practically

impossible to standardize civil structures given the wide spectrum of loading combinations that
are to be applied according to equipment, seismicity, wind conditions, transportation etc. In this
context, standardizing a structure means to make it easily adaptive (versatile) to loading conditions
of increasing level, e.g. by simply adding some structural elements and/or modifying a limited
number of existing ones.
3. Improve functionality
All the items in the “indirect benefit” list can be seen as contributions to the functionality of the
construction, i.e. the capability of fulfilling, in a more efficient and economical way, the needs for
which it has been designed.
3.2 General issues
A single case study has been analysed in this first year of activities, regarding a typical pipe rack
structure (whose structural layout is depicted in Figure 3.1): the main findings and proposals
resulting from the analysis will be summarised in the Appendix. Here, however, an attempt is
made to draw some general considerations for a wide class of industrial buildings, i.e. open steel
frames carrying equipment, characterised by:
» rectangular, or close to, structural plan, often showing significant elongation;
» need for an open transversal section;
» irregular vertical spacing of horizontal beams;
» lack of well-defined horizontal levels, both for the absence of flooring systems and for the
vertical offset between beams running in the two directions;
» lack, in many cases, of an efficient horizontal bracing system connecting the vertical frames;
» strict requirements for fire resistance;
» high degree of transparency against wind actions.
3.2.1 Weight reduction: welded joints vs bolted joints
As already mentioned in the introductory remarks, the structural optimization for weight reduction
is often prevented by strict requirements, connected to stiffness and/or bearing capacity of the
considered structures. Nevertheless, some improvement in the overall weight can be achieved by

Figure 3.1 - Structural layout of the case study considered in this chapter: general
view (top), longitudinal frame (left bottom), transversal frame (right bottom)
considering welded joints instead of bolted connections.

Welded joints are obtained by suitably combining several welds on different parts of the structural
elements. By comparing welded and bolted joints, one finds that the former show several advan-
tages: 1) "natural" monolithic behavior of the joints; 2) higher stiffness with a limited adoption of
additional members; 3) simpler layout, with the consequence of additional freedom in the
structural design. All these features might entail a reduction of structural weight, particularly
because connecting plates and packing plates, which are commonly adopted in bolted connections,
are not necessary in welded joints. As a rule of thumb, in the case of steel frames like the module
considered herein, the impact of plates and bolts on the overall weight may easily reach 10%.

On the other hand, welded connections are characterised by some critical issues, mainly connected
to the possible presence of defects (cracks, lamellar tearings, inclusions, etc.). For this reason, it
is compulsory to investigate the accuracy of welds by means of non-destructive techniques. It
seems that the testing task can be reasonably handled in the case of modular structures, which
are mostly assembled in the workshop: in such a controlled environment, non- destructive analyses
can be carried out in an easier way.
3.2.2 The determination and treatment of wind actions
The design of the structure in the case-study module against wind effects was performed within
the framework of Eurocode 1- part 4 [8]. Accordingly, loads are given as static forces depending
on the site design wind (average velocity and turbulence), on the system aerodynamics and on a
structural coefficient. No attempt was made to reduce loads working on the first two aspects;
coming to the structural coefficient we recall that it takes account of the dynamics effects, which
increase the response, and of the non-simultaneous occurrence of peak pressures over the
exposed structure, which has a beneficial effect.
The choice of a unit coefficient, which is an usual option and was made in the original design, is
based on the assumptions that the two effects cancel each other; thus, in order to reduce the
wind loading, the dynamic behaviour must be improved and/or spatial correlation effects, re-
sponsible for non-simultaneous pressure peaks, must be better exploited.
For usual structures, the first aspect is related to the increase in lateral stiffness, which was a
general objective of the case study. In addition, some work has been done on the spatial correlation
effects, both on the structural coefficient approach and by a more refined technique based on
complete dynamic analysis. Some encouraging results have been obtained but it is deemed that
the topic could deserve a more substantial research effort; in this perspective the performance of
a test campaign in the wind tunnel could be evaluated.
3.2.3 Stiffness: horizontal bracing as a prototype problem of code application
The introduction of horizontal bracing, at least on top of the structure, improve the overall structural
behaviour in many respects, leading to a more effective collaboration among transversal frames
which results, in turn, on a favourable internal force redistribution for the cases of transversal
loading (e.g. wind or PSV operation). In addition, it must be quoted that Eurocode 8 Part 1 [9]
states, among the “Basic principles of conceptual design”, the following:
4.2.1.5 Diaphragmatic behaviour at storey level
(1) In buildings, floors (including the roof) play a very important role in the overall seismic behaviour of the
structure. They act as horizontal diaphragms that collect and transmit the inertia forces to the vertical

structural systems and ensure that those systems act together in resisting the horizontal seismic action. The
action of floors as diaphragms is especially relevant in cases of complex and non-uniform layouts of the
vertical structural systems, or where systems with different horizontal deformability characteristics are used
together (e.g. in dual or mixed systems).
The “floor action” is one of the numerous problems which we have encountered while attempting
to apply standard code provisions, which are targeted to usual buildings (office, residential, ...),
to structures like the one here considered; in some cases we found the provisions simply impossible
to be fulfilled. This is the case of horizontal diaphragms or bracing systems, which should be in-
troduced at all levels where horizontal beams develop a load carrying action; this is clearly
impossible in a pipe rack of the type here considered, both for the difficulty of defining the
relevant levels (longitudinal and transverse beams run at different heights) and for the almost
sure interference with the piping layout.
In situations of this type we tried to apply the principle more than the specific rule; in the case of
conflicting issues, as for vertical bracing systems, we chose to satisfy what we regarded as the more
important aspect; the idea is, in future developments of the research, to validate the design choices
here adopted, sometimes violating code provisions, by performing refined non-linear analysis of
some prototype structures under dynamic actions due to seismic events of different severity.
Coming back to the case study, we have introduced a horizontal bracing system only at the top of
the structure, this being the most efficient position on structural grounds and an almost trouble-
free option in terms of compatibility with equipment.
3.2.4 Stiffness 2: composite steel-concrete columns and the versatility concept

In the case study wide-flange laminated columns were encased in a concrete fire protection at the
first floor. Concrete was regarded as non-structural given its high probability to be removed or
heavily damaged for connecting pipes and pieces of equipment to the steel element. In this study
it was proposed to investigate the use of hollow steel sections filled with reinforced concrete on
site, with the following advantages:
» concrete stiffness and strength is exploited;
» weight at transportation is reduced;
» concrete pouring is easy;
» equipment connection to the columns is easier.

The proposal must be obviously checked in terms of fire resistance; the internal concrete provides
beneficial thermal inertia, but an external protection could be necessary for higher performance.
In addition the development of standardized joint connections between tubular columns, laminated

beams and braces should be pursued.

The use of steel hollow sections can be seen as a first attempt to introduce the versatility concept;
the bare section, and its connections to adjoining elements, can be introduced as the basic
structural solution. Its performance can be upgraded in terms of structural and fire resistance just
by adding internal concrete and/or external protections, thus minimising the modifications to the
overall design.
3.2.5 Seismic design strategy for Ultimate Limit States

The seismic design strategy strongly depends on the behaviour factor (q factor in Eurocode 8),
which allows for the reduction of the internal forces computed through a linear analysis, accounting
for the beneficial effect of the “ductility of the structural system”. The latter term is to be intended
in a broad sense, covering both a stable and reliable hysteretic behaviour at the material/element
level and a favourable global structural dynamic response in the inelastic range.
Behaviour factors and regularity
When weight reduction is assumed as the basic design target, the pursuit of the highest possible
structural factors seems an obvious choice. This, however, would imply to meet rather strict
regularity conditions, in terms of both plan layout and variation of structural properties (mass,
stiffness, resistance) with elevation. The latter aspect, i.e. regularity in elevation, conflicts with the
necessity of being free to place equipment-carrying elements (i.e. beams) at irregularly spaced
vertical positions; for this reasons, a reduction factor of 0.8 has been accepted in applying Eurocode
8 provisions to account for lack of regularity.
Behaviour factors and local ductility

Another fundamental aspect affecting behaviour factors at the local level is the section geometry,
which affects phenomena like local buckling. In this respect compact sections or laminated section
having small width-to-thickness ratios are rewarded by all seismic codes. In Eurocode 8, high and
medium ductility structures, with remarkable difference in the q factors, are defined according to
section properties: in the case-study example, beams and columns sections met the conditions
for the high ductility. Nevertheless, it was decided to treat the structure as a medium ductility one;
this was seen as a compromise solution, accounting for other code provisions that cannot be
fulfilled by the design. It must be also noted that when very large behaviour factors are introduced,
as it happens for high ductility structures, seismic design is prone to be governed by serviceability
conditions, which must be checked in the elastic range and do not benefit from the q value;
therefore, increasing the latter does not automatically imply a reduction in element size/weight.
Bracing systems and behaviour factors
An additional aspect that has been addressed in the case study is related to the so called “Chevron”
(inverted V) braces, which are rather popular in the design of industrial structures even though in

Eurocode 8 they are penalized by a low value of the behaviour factor, equal to half the one
granted to traditional X bracings; such low value is consistent to the design approach implicit,
though not declared, in the code. For this reason, an alternative design approach has been
followed herein, which justifies the use of the same behaviour coefficient as for X braced systems;
this is achieved at the cost of modifying the design of the top beam, to which the diagonal braces
are connected, in order to increase its resistance and lateral stability. This can also contribute to
the versatility goal, since the modification of few elements leads to a significant upgrade of the
seismic performance of the system.
3.2.6 Seismic design strategy for damage limitation
Seismic design of steel structures is often conditioned by the damage limitation state (DLS) for
frequent events, i.e. having a return period of the order of the service life, rather than to resistance
and ductility under the strong motion, characterized by a much lower probability of occurrence.
This is the typical design situation for moment resisting frames (MRFs), which are intrinsically de-
formable and thus prone to non-structural damage in serviceability conditions, but ensures very
favourable design coefficients against ultimate limit state (ULS) seismic actions.
A similar situation occurred, in the case study here performed, when the design of transverse
frames against high-level seismic forces was addressed; it was found that the MRF resistance
fulfilled the requirements of Eurocode 8 against ULS, while was too deformable to satisfy no
damage requirements. A non-conventional design solution was proposed (following [11]), based
on the introduction of a bracing system whose contribution was accounted for stiffness, within
the context of DLS, but disregarded in ULS checks regarding resistance and ductility. This solution,
allowing for the use of slender braces and lighter joints, can be seen again as an application of
the versatility concept.

4 | Conclusions
This work presented the results of a wide spectrum research on plant modularisation. The three
objectives of the study were to:
» compare concepts expressed in literature and actual practices
» assess these practices in order to identify gaps to fill for an enhanced modularisation appli-
cability;
» analyse some of the identified technical criticalities in order to boost

the development of general and on hand design solutions.
In order to fulfil this threefold purpose a survey was conducted among a significant sample of the
Italian EPC project delivery chain. The survey’s results are disclosed in chapter one that consists of
four sections.
Section one presented the literature review, with particular reference to drivers and criticalities
generally associated with modularisation. Section two explained the questionnaire structure hi-
ghlighting methodological aspects, enlisting the companies involved in the survey and specifying
the interviewed roles.
Section three showed and analyzed in depth the interviews results. The first of them is a new de-
finition of modularisation consistent with the actual practices of a significant sample of the Italian
EPC supply chain. Furthermore an identification and a semi-quantitative prioritisation of modula-
risation driving factors and constraints was obtained. Schedule, Site conditions and Labour related
issues showed to be the most important drivers for modularisation. On the other hand Modules

Conclusions
Engineering, Module transport and the higher Complexity associated to modular projects are
considered major constraints to the choice of this construction strategy.
Section four, on the basis of the survey results, identified several improvement areas in modula-
risation management. The aim is to provide ideas on the way of the definition of a new management
framework for modular plants projects, since standard management approaches showed to effec-
tively not address this kind of projects. The study highlighted that the hinges of such a management
framework should be prompt decision making tools able to cope with uncertain and scarce infor-
mation as well as proper organisational structures and supervision tools. Procurement and engi-
neering activities should be customised for modularisation, considering every existing feed-back
loop from the early stages of the project, maximising modules’ functional completeness and mi-
nimising the reduction in components commonality due to the decoupling of procurement activities
for different modules. Also new contractual forms customised for modular plant projects was pin-
pointed as a main need in order to avoid conflicts and maximise the benefits for all the stakeholders
involved in the project.
As a second purpose of this report some engineering aspects related to mudule handling and
transportation and to their structural analysis and design are discussed. In this light, in chapter
two an overview of the problems criteria and solutions is presented in relation to the procedures
for module sea transportation and land transportation by means of SPMTs (Self Propeller Modular
Trailers): attention is devoted to
» the definition of the sea motion, related to the safety of fastening and grillage procedures,
and to the resistance of the modules;
» the choice of the vessel and the verification of its strength and stability;
» the safety and functionality of loadin and loadout operations, including considerations
about the effect of the tidal range, the characteristics of the quay and the SPMT operation;
» the configuration of the hydraulic system supporting the wheel boogies of the SPMT;
» the resistance and stability of the SPMT.

Finally, chapter three has been devoted to structural analysis and design of modules; an attempt
to draw some general consideration has been made, based on the results of a case study regarding
a real life pipe rack and described in detail in the Appendix.
These consideration address two fundamental issues; the first points to the necessity of developing,
for each identified class of modules, standardized design criteria and procedures, leading to the
definition of what we called “versatile” solutions. In these solutions, a basic structure, suitable for
resisting favourable loading conditions (tipically seismic, wind and lifting/transportation), can be
upgraded to more demanding situation by adding or modifying a few structural elements.

Conclusions
As a second remark, we can observe that such standardized procedure would strongly benefit
from the development of a “code of practice”, based on the provisions of an international code
(e.g. the Eurocodes) and aiming to extract from the code itself a selection of suitable and standar-
dized design options. This activity could also lead to some proposal for interpreting or modifying
code provisions, in order to meet the characteristics of the modularized structural systems;
obviously, these proposals should be strongly supported by a RD activity, in order to demonstrate
their feasibility and validity. A typical example should be the performance of refined non-linear
dynamic analyses for supporting some innovative proposals in the field of seismic design.

References
[1] K. H. Bae, H. C. Kim, M. H. Chang and S. K. Sim, “Safety evaluation of the inherent a n d
passive safety features of the smart design,” Annals of Nuclear Energy, no. 28, pp. 333-349, 2001.
[2] S. M. Goldberg and R. Rosner, “Nuclear Reactors: Generation to Generation,” AMERICAN
ACADEMY OF ARTS & SCIENCES, Cambridge, (MA,US), 2011.
[3] M. Carelli, L. Conway, L. Oriani, B. Petrovic, C. Lombardi, M. Ricotti, A. Barroso, J. Collado, L.
Cinotti, N. Todreas, D. Grgic, M. Moraes, R. Boroughs, H. Ninokata, D. Ingersoll and F. Oriolo, “The
design and safety features of the IRIS reactor.,” Nuclear Engineering and Design, no. 230, p. 151–
167, 2004.
[4] S. Armenante, D. Rivellini and L. Ferrazzi, “Modularisation in onshore LNG base-load Plants,”
Impiantistica Italiana, no. 5, pp. 42-55, 2013.
[5] R. Brookfield and C. Jeremy, “Modularisation: a pioneering approach,” The Petroleum Eco-
nomist, pp. 1-3, 2011.
[6] C. Caswell, C. Durr, H. Kotzot and D. Coyle, “Current Myths About LNG,” KBR, Houston, 2013.
[7] P.H. Jameson, “Hydrocarbon Processing,” Tyler, TX (USA) , 2007.
[8] EN 1991-1-4:2005, “Eurocode 1 - Actions on structures - Part 1-4: General actions- Wind ac-
tions,” 2005.
[9] EN 1998-1:2004, “Eurocode 8: Design of structures for earthquake resistance – Part 1:
General rules, seismic actions and rules for buildings,” 2004.
[10] EN 1990:2002/A1:2005, “Eurocode - Basis of structural design,” 2005.
[11] L. Martinelli, M.G. Mulas and F. Perotti, 'The seismic behaviour of steel moment-resisting
frames with stiffening braces', Engineering Structures, Vol. 20, No. 12, pp. 1045-1062, 1998.

APPENDIX A: Questionnaire
Interview on modularisation management
Interviewee: ..........................................................................
Position within the company:.................................................
Company: .............................................................................
Date:......................................................................................
Interviewer: ..........................................................................
Introduction

Appendix A
Questionnaire
I. Definition of modularisation
a. What do you understand by the term modularisation? Can you provide an example on this?
Definition Example
b. In your opinion, which of the following definitions best expresses the concept of modularisation/
modularity? Can you rank these?
Definition of modularity Ranking (1-4)
“Modularity is a general systems concept; it is the extent to which a simple or complex industrial
product can be broken down and reassembled.” [1]
[...] the definition of modularisation is: decomposition of an industrial product into minor or
major parts or blocks(modules) , fabricated and assembled offsite ,including , to the maximum
extent, complete or partial systems, moving offsite productive MHRs. [2]
“Modularity [...] is a bundle of characteristics that define

(a) interfaces between elements of the whole,
(b) a function-to-component [...] mapping that defines what those elements are, and
(c) hierarchies of decomposition of the whole into functions, components, tasks etc.” [3]
“Product modularity is a systems design strategy that can be used to:

1) manage complexity by hierarchically decomposing a whole into parts and by mapping
functions to parts in order to minimise interdependencies, to thereby enable the pursuit for
2) economies of scale by standardising such parts and
3) variability through standardised interfaces that allow the use of interchangeable such parts,
or
4) other such benefits.” [4]

Appendix A
c. Considering the following areas of interest in respect of modularisation, have you had any
specific experience of them in your professional career? Can you provide examples for the cases
presented?
Num. Areas of application Importance
1 Product
2 Plant
3 Yard
4 Intangible product: software
5 Knowledge & Capabilities
6 Service
7 Organization
8 Function
9 Documentation
a. In your professional opinion, who are the main decision makers about the need of modulari-
sation? For what reasons?
Num. Internal Actors Main reasons
5
Num. External Stakeholders Main reasons

Appendix A
b. Considering your professional experience, which “actors” usually advocate against modulari-
sation? What are their arguments and what are their motives?
Num. Internal Actors Main arguments Main motives

1
2
3
4
5
Num. Internal Actors Main arguments Main motives
1
2
3
4
5
II. Driving factors and objectives

a. Can you list the main factors driving the modularisation? Please assign a score between 0 and
100 according with the estimated importance of each driver.
Num. Drivers Importance
10

Appendix A
b. Can you make an example for each of the mentioned “Drivers”?
Num. Examples of Drivers
1
2
3
4
5
6
7
8
9
10
c. In your opinion, which of the following purposes best express the objectives
associated to modularisation for each stakeholder? Please assign a score bet-
ween 0 and 100 according with the estimated importance of each purpose.
Num. Purposes of modularisations IMPACT
EPC Service Utility Manu-
Provider facturer
1 Balance standardization with customization

2 Reducing project cost
3 Reducing project delivering time
4 Reducing complexity
5 Reducing risk
6 Reducing design and/or manufacturing efforts
7 Enhancing project manageability
8 Enhancing market competiveness (e.g. better services)
9 Enhancing innovation
10 Increasing safety
11 Increasing construction efficiency
12 Reducing and shortening of MPW peak (on site)
13 Reducing imported MPW
14 Mitigating the lack of lay-down areas constraints
15 Mitigating social impacts at site due to construction
16 Increasing sustainability of the site works

Appendix A
III. Modularisation constraints

d. Can you list what you think are the most critical issues (e.g. barriers, constraints) associated
with modularisation? Please assign a score between 0 and 100 according with the estimated im-
portance of each issue.
Num. Issue Importance

1
2
3
4
5
6
7
8
9
e. Please provide examples for the mentioned barriers and constraints?
Num. Examples of barriers and constraints

1
2
3
4
5
6
f. According with your experience, in which way do you remove or limit the effect of the barriers
7
and constraints to the use of modularisation?
Num. Methods to remove or limit barriers and constraints

1
2
3
4
5
6
7
8
9

Appendix A
Note

APPENDIX B: Case study
B.1 Analysis of the structural response to wind loads
In serviceability conditions, the modular structure considered in this case-study is subject to the
external loads summarized in the following table.
CATEGORY LOADS
Permanent, structural Self weight, Dead load
Permanent, non-structural Fire proofing, Piping weight
Anchor forces, PSV,

Accidental Operating loads, Wind & snow,
Seismic load
Table B.1: loading conditions
It is worth noting that the considered classification is not intended for the case of transient
situations. For instance, the transportation phase may involve severe loading conditions, which
deserve a thorough treatment and is the subject of ongoing research.
Since the aim is to achieve weight saving via load reduction, it is reasonable to disregard permanent
actions (which are normally established by the operative requirements) and to consider a detailed
analysis of accidental loads. More specifically, focus will be on wind and seismic actions, which
are the heaviest loads in many practical situations.
B.1.1 Wind action: the structural factor
By considering the Eurocode 1, part 1.4, on finds that the wind actions can be represented by the
following static equivalent forces:
Fw = cs cD cf qp (zs) Aref

Appendix B
qp is the peak velocity pressure, which in turn depends on the wind velocity and the exposure co-
efficient at height zs. Aref is the reference area, which is properly modified by the force coefficient
cf : the latter is usually obtained by the code or from literature data.
This section contains the detailed study of a sensible coefficient, namely the structural factor cs cd
, which is precisely defined in the Eurocode 1 Part 1.4:
6.1 General
(1) The structural factor cscd should take into account the effect on wind actions from the non-simultaneous
occurrence of peak wind pressures on the surface (cs) together with the effect of the vibrations of the
structure due to turbulence (cd).
The aforementioned Standard provides a detailed procedure for the numerical evaluation of the
structural factor. The detailed description is reported in the Annex B of Eurocode 1 Part 1.4 and it
is not reported in this document for the sake of brevity. Some general information are usefully re-
minded: 1) the non-simultaneity coefficient cs is basically dependent on the turbulence features
and on the structural dimensions; 2) the latter are referred to parts of the structure which simul-
taneously and independently supports the wind action; 3) the dynamic coefficient cd depends
also on the dynamic behaviour of the structure, which affects some specific aerodynamic coefficients;
4) the dynamic features are assumed to be summarized by the properties of the fundamental vi-
brating mode (frequency, shape and damping factor).
It is worth noting that such a procedure has been developed for some specific cases (i.e. slender
civil buildings, horizontal structures, pointlike structures) which do not encompass the considered
structure. Nevertheless, the procedure has been applied and a specific validation has been
provided by dynamic analyses.
B.1.2 Structural factor for the original layout
As a first step, the original layout of the pipe-rack has been considered. The modal analysis
yielded a set of closely spaced modes, which are referred to single transversal frames. The situation
is depicted in Figure B.1, which contains the plan view of the first three vibrating modes: it is clear
that, in the absence of specific horizontal bracing, each single frame vibrates independently with
respect to the adjacent frames.
Taking this structural behavior into account, the non-simultaneous action of wind should not be
considered on the whole structural assembly but, more realistically, on each single bay connected
to the vertical frame. The structural factor is thus computed by considering the frequency of the
first natural mode and a structural width equal to the frame spacing (namely, b = 6 m). The com-
putation of the structural factor is summarized in the following tables, where the naming convention
is in agreement with Eurocode 1.

Appendix B
Figure B.1 - Plan view of the first three vibration modes for the original layout of
the pipe rack, in the absence of horizontal bracing; the frequency and the period of
vibration are indicated for each mode.
The computation yields a structural factor which is considerably larger than one: the nominal
wind action on the pipe rack should be increased by about 28% in view of the unsatisfactory
dynamic behavior of the structure. A plain provision in order to reduce the wind action is
represented by the introduction of some structural connection between the transverse frames.
B.1.3 Modified layout: horizontal bracing system
In the previous section, the structure has been considered to be composed by four independent
transverse frames, in the absence of specific connecting elements. This situation is clearly harmful
for the structural factor; moreover, further drawbacks are represented by the structural response
to random and non-simultaneous loads (connected for instance to the Pressure Safety Valves)
and by the interaction with the technological equipment. The former issue will be treated at the
end of this section. Now, it is worth spending some words on the fact that the structure is designed
with the purpose of containing several pipe lines, which may run in the longitudinal direction. In
principle, such elements could provide a certain constraint for the relative movement of adjacent
frames. As a matter of fact, the common tendency in the design procedure is to consider the pipes
as non-structural elements, with the purpose of avoiding additional stress which may endanger
the plant operation. In this sense, the presence of horizontal bracing could represent a beneficial
provision because it removes the loads acting on the pipes due to relative displacement of the
transverse frames.

Appendix B
Wind load analysis Structural factor

Fundamental value of wind velocity [m/sec] vb,0 = 26.2 Reference height [m] zs = 10.2
Directional factor cdir = 1 Roughness factor cr (zs) = 1.01
Season factor cseason = 1 Orography factor c0 (zs) = 1
Basic wind velocity [m/sec] vb = 26.2 Mean wind velocity at height zs [m/sec] vm(zs) = 26.474
Height of the building [m] h= 17 Turbolence intensity at height zs Iv (zs) = 0.188
Terrain category II Basic velocity pressure at height zs qb(zs) = 438.03
Roughness length [m] z0 = 0.05 Peak velocity pressure at height zs [N/m ]2

qp(zs) = 1014.6
Minimum height [m] zmin = 2 Building width [m] b= 6

Terrain factor kr = 0.19
Reference scale lenght [m] Lt = 300
Roughness factor cr (he) = 1.01
Reference height [m] zt = 200
Orography factor co (he) = 1
Coefficient [m] α= 0.52
Mean wind velocity at height he [m/sec] vm(he) = 26.47
Turbulence lenght scale [m] L(zs) = 63.79
Turbulence factor kl = 1
Background factor B=
2
0.679
Standard deviation of the turbulent σv = 4.927
component of wind velocity Duration of the design event [s] T= 600
Turbulence intensity Iv (he) = 0.188 Natural frequency of the building [Hz] n1x = 0.901
Exposure factor (1+7Iv) ce(he) = 2.316 Non dimensional frequency fL(zs,n1x) = 2.171
Air density [kg/m3] ρ= 1.25 ηh = 2.661
Mean velocity pressure at height he [N/m ]2

qb(he) = 438 ηb = 0.939
Peak velocity pressure at height he [N/m2] qp(he) = 1015 Loss of correlation factor (vertical) Rh = 0.30549
Loss of correlation factor (horizontal) Rb = 0.5845

Table B.2 - Wind load analysis table
Non dimensional power spectral density function SL = 0.07854
Structural damping (logarithmic decrement) δs = 0.05
Aerodynamic damping (logarithmic decrement) δa = 0.00033
Damping due to special devices δd = 0
Total damping (logarithmic decrement) δ= 0.05033
Resonance response factor R=

2
1.37498
Up-crossing frequency ν= 0.73723
Peak factor kp = 3.66247
Structural factor cscd = 1.2839
Size factor cs = 0.8999
Dynamic factor cd = 1.4267
Table B.3 - Structural factors table

Appendix B
After several preliminary studies, it has been obtained that the desired target can be reached by
considering a single bracing system, on the top level of the pipe rack. This layout is endowed with
several positive features: 1) structural efficiency; 2) minimum interference with piping system; 3)
full compatibility with the common pipe rack arrangement (staggered horizontal beams in the in-
termediate floor, not on the top level). The modified layout is depicted in Figure B.2, which shows
the cross-bracing system constituted by simple T-shaped steel elements.
Figure B.2 - Modified layout, with horizontal cross-bracing on the top level.
With this simple provision, which entails a negligible addition of material, the dynamic behavior
is substantially changed: all the transverse frames are now involved in the first vibration mode,
characterized by a natural frequency which is slightly higher with respect to the original case.
Figure B.3 shows the first modal shape (transverse view and three-dimensional view).
In view of the cooperation of the four transverse frames, the structural factor can be tentatively
computed by changing the reference width, which should be chosen as the overall width of the
pipe-rack. In this way, both the size factor and the dynamic factor are reduced, finally obtaining a
structural factor equal to 1.10 (-14% w.r.t. the original value).
As anticipated, the presence of a horizontal bracing system is also advantageous for the loading

Appendix B
Figure B.3 - Shape of the first vibrating mode, in the presence of the horizontal
cross-bracing system; natural frequency of the considered mode: 0.93Hz.
condition represented by the release of Pressure Safety Valve (PSV). PSVs, which are usually con-
centrated on the top floor of the pipe rack, automatically release the fluid from the piping system,
when the pressure (or temperature) exceeds preset limits. The sudden release of fluid involves a
dynamic effect, which, for design purposes, can be represented by concentrated loads in the
transverse direction. The effect is essentially impulsive, as its duration is by far shorter than the
structural fundamental period. It should be noted that the loading condition is mainly related to
the first transverse mode; by neglecting the effects of damping, which are of minor importance for
the evaluation of the response peak, the single-mode oscillations due to an impulsive effect are es-
sentially harmonic. The combination of the effects of various PSVs is thus coincident with that
among harmonic responses sharing the fundamental frequency. If one considers the phases, the
main outcome is that the precise timing of the PSV activation is not predictable; however, for te-
chnological reasons, the simultaneous activation of the valves is highly unlikely. Assuming that the
above phases are uncorrelated, a generic mechanical effect R on the structure can be computed by
combining the effects of each single PSV Ri by means of the SRSS formula (Square Root of the Sum
of Squares):
R= √ ∑R i i
2
It is quite easy to realize that, in the case of independent frames, there is no difference with respect
to the standard computation for simultaneous application of the PSV load. In fact, the PSV action is
just balanced by a single frame, with no cooperation of the adjacent ones. Conversely, the presence

Appendix B
of the horizontal brace is beneficial both for the participation of the whole structure and for the
further reduction due to non-simultaneity of the external load. A numerical comparison is provided
by considering the bending moment in the most loaded frame, for the PSV action. In the case of in-
dependent frames, the maximum bending moment is 8.31 kNm; conversely, by applying the SRSS
superposition in the case of horizontal brace, one obtains 3.39 kNm. A 60% reduction of bending
moment in the transverse frame is hence obtained.
B.1.4 Further modification: stiffened columns
As it has been stated in the introductory Section, the structural stiffness can be also increased by in-
troducing a modification to the column design, which is usually based on rolled section steel
profiles protected by non-structural concrete as a fire-proofing provision. As a promising alternative
the adoption of hollow sections filled by poured concrete has been here considered. This modification
entails an increase of stiffness in view of the possible collabouration between steel and concrete
elements. Note that the concrete covering of standard I-shaped beams is exposed to the risk of in-
situ demolition and reconstruction for plant requirement which may intervene during the structure
lifetime. For this reasons, it is common practice to neglect the presence of concrete from the
structural point of view. Conversely, in the case of concrete inside hollow sections, such a risk is
completely eliminated and, by introducing some suitable steel- concrete connection, a composite
cross-section can be considered in the structural analysis. The lateral stiffness is thus increased,
with the consequence of more favorable dynamic response to the wind action. Clearly, the compu-
tation of cross-section inertia should comply with the standard restrictions. For instance, in order to
take into account the non-perfect adherence of the heterogeneous materials and the concrete
cracking under tensile load, a corrective coefficient (lower than unity) should be introduced when
computing the concrete contribution.
The analysis has been carried out by considering both the horizontal bracing system and the
modified columns (see Figure B.4). The latter are filled with concrete until the level +4.68 m, which
correspond to the first floor level.
The fundamental vibrating mode of the modified structure is quite similar to the previous case,
and, for this reason, no graphical representation is reported herein. Conversely, there is a significant
variation of the fundamental frequency which now attains the value 1.28 Hz. The structural factor
has been computed in this situation and a value close to unity is obtained, in view of a substantial
reduction of the dynamic factor. It is finally possible to conclude that an increase of the structural
stiffness, given by a horizontal bracing system and a modification of the columns,finally yields the
desired compensation between dynamic amplification and non-simultaneous effects.
B.1.5 Validation via dynamic analysis

The structural factor has been computed on the basis of analytical expressions provided by the Eu-

Appendix B
Figure B.4 - Modified layout, with horizontal cross-bracing on the top level and
hollow columns filled with concrete until the first floor level (i.e. +4.68 m above the
foundation level).
rocode 1, Part 1.4. The application of such formulas is restricted to some simple structural cases,
representing, on aerodynamic grounds, point-like or slender bodies, excluding extended structures
as the pipe rack considered herein. It must be said that the aerodynamic behaviour of a typical
pipe rack should deserve detailed experimental investigation in order to characterize the pressure
distribution on the structural and equipment components. Here, the application of the simplified
procedure of Eurocode 1 has been validated in terms of structural behaviour by performing a
critical comparison with respect to a complete dynamic analyses. In particular, a step-by-step
dynamic analyses was performed, characterized by the application of forces whose space-time di-
stribution is consistent to the stochastic model of turbulence which is the basis of the definition of
the dynamic coefficient.
The analysis is based on the same hypotheses as adopted in the simplified procedure, i. e. the de-
finition of the action of the wind as a superposition of two effects: a static force due to the mean
velocity and a dynamic one due to the fluctuations of the velocity and aerodynamic damping. The
latter component of the dynamic force is neglected, while the former (due to wind fluctuation) is
represented as a set of dynamic concentrated forces that are directly applied to the model, along
with the static components. The following assumptions apply :
» only the effects of drag forces is considered (neglecting the lift )

Appendix B
» the transverse component of turbulence is neglected The static

component of the wind force is defined as:
1
Fj,sta = ― ρ Aj CF ,j W 2 ( zj )
2
where ρ is the air density, Aj is the surface area exposed at node j, CF ,j is the coefficient of aero-
dynamic force, W ( zj ) is the value of the average speed at level zj .
The dynamic component of wind force is defined as:
Fj,dyn = ρ Aj CF ,j W ( zj ) w ( zj ,t )
where w ( zj ,t ) is the component in the direction of the wind fluctuation due to turbulence at level
zj. For the determination of this component, it is necessary to generate the time history of the fluc-
tuating velocity, through a stochastic model of wind fluctuations. The model takes into account,
via the space coherency function, the correlation of the turbulent components in the different ap-
plication points. Each generated time history of the fluctuation corresponds to a dynamic component
of the wind force, that is superimposed to the static component of the point in question .
Among the outputs of the analysis, for the purpose of comparison, the shear force at the base of
each column is considered. The analysis has been referred to the modified layout, in the presence
of horizontal bracing systems and stiffened columns. The peak force has been extracted from the
time history of shear forces, obtained by means of the dynamic analysis and summed to the static
component; the comparison has been carried out with respect to the analogous force, that is
statically computed by applying the wind action modified by the structural factor. In this way, the
result of the more refined analysis is compared to the static-equivalent analysis and the procedure
for obtaining the structural factor can be validated. The outcomes of the comparison are reported
in the next table.
Good agreement between the two analyses is obtained. A maximum discrepancy of 32% is
obtained on column 6, but the most important thing is that the static-equivalent analysis is on the
safe side. The maximum force, which is normally used in order to design all the columns, is
caught with excellent precision (3% error) by the static-equivalent analysis.
From the above results, it is possible to conclude that the procedure reported in Eurocode 1 for
the structural factor can be successfully applied to this kind of structure, provided that a validation
is performed in terms of aerodynamic behavior.

Appendix B
Shear force (kN)

Column n.
Dynamic analysis Static-equivalent analysis
1 33.06 31.15
2 34.15 35.64
3 36.93 35.65
4 36.93 32.50
5 26.97 31.15
6 26.98 35.64
7 27.54 35.65
8 27.50 32.51
Table B.4 - results of wind analysis
B.2 Seismic analyses
The achievement of an earthquake-resistant structure involves the examination of several specific

issues. Most of them are related to the capability of the structural system to dissipate the kinetic
energy due to ground shaking, thus avoiding the catastrophic collapse at the price of severe pla-
sticization in specific elements. Moreover, a proper seismic design should also entail an acceptable
behavior in exercise conditions, for the cases of earthquake endowed with lower severity and
shorter return time: in general, the serviceability conditions for steel structures are related to the
limitation of horizontal relative displacements between adjacent stories.
The purpose of this Section is to provide some specific hints for the design of modular systems in
seismic areas. Two possible scenarios are considered: moderate values of PGA (peak ground ac-
celeration), around 0.2 g; severe earthquake conditions, with PGA around 0.4 g. In this way, it will
be possible to highlight different structural concerns, both for the ultimate limit state and for the
exercise conditions. As a preliminary step, the problem of the correct evaluation of the behavior
factor is thoroughly treated. Then, the main results for the seismic analyses of the case- study (the
same pipe-rack as for the previous Sections) are reported in order to introduce the proposed
structural modifications and their usefulness.

Appendix B
B.2.1 Proper evaluation of the behavior factor
The first step in a seismic analysis consists in the definition of the behavior factor (called q in the
Eurocode 8). This factor is involved in the definition of the design actions in case a linear dynamic
analysis is carried out as an approximation of the truly nonlinear behavior of the structure (of
course, if the Ultimate Limit States are considered); in other words, the behavior factor allows
one to operate in an easy framework by performing the analysis in the elastic field and then by re-
ducing the results, in terms of internal actions, in order to take account of the dissipation due to
inelastic phenomena.
According to Eurocode 8, 3.2.2.5 “the behavior factor q is an approximation of the ratio of the seismic
forces that the structure would experience if its response was completely elastic with 5% viscous
damping, to the seismic forces that may be used in the design, with a conventional elastic analysis
model, still ensuring a satisfactory response of the structure”. In particular, the behavior factor defines
the design spectrum starting from the elastic response spectrum; with the exception of the first linear
branch, the design spectrum is obtained from the elastic one through division by q:
MSa (T,ζ )
FS,d =
q
The determination of the behavior factor is influenced by the properties of the structure (symmetry
and regularity) and by the ductility at local and global level. Special attention should be paid to
the study of the regularity of the structure under consideration. Eurocode 8, 4.2.3.1, reads:
(1)P For the purpose of seismic design; building structures are categorised into being regular or non-regu-
lar.
(2) This distinction has implications for the following aspects of the seismic design:
» the structural model, which can be either a simplified planar model or a

spatial model;
» the method of analysis, which can be either a simplified response

spectrum analysis (lateral force procedure) or a modal one;
» the value of the behaviour factor q, which shall be decreased for buildings
non-regular in elevation.
Table B.5 indicates the effects of structural regularity on the seismic analysis and design; it is
worth noting that, if the requirement is not met, the behavior coefficient must be penalized. The
requirements for regularity in elevation are summarized in section 4.2.3.3 of Eurocode 8:

Appendix B
Regularity Allowed Simplification Behaviour factor

Plan Elevation Model Linear-elastic Analysis (for linear analysis)
Yes Yes Planar Lateral force a Reference value
Yes No Planar Modal Decreased value
No Yes Spatial b Lateral force a Reference value
No No Spatial Modal Decreased value
a
If the condition of 4.3.3.2.1(2) a) is also met.
b
Under the specific conditions given in 4.3.3.1(8) a separate planar model may be used in each horizontal
direction, in accordance with 4.3.3.1(8)
Table B.5 - Consequences of structural regularity on seismic analysis and design
Criteria for regularity in elevation

(...)
(2) All lateral load resisting systems, such as cores, structural walls, or frames, shall run without interruption
from their foundations to the top of the building or, if setbacks at different heights are present, to the top
of the relevant zone of building.
(3) Both the lateral stiffness and the mass of the individual storeys shall remain constant or reduce
gradually, without abrupt changes, from the base to the top of a particular building.
(4) In framed buildings the ratio of the actual storey resistance to the resistance required by the analysis
should not vary disproportionately between adjacent storeys.
(...)
The case-study does not meet the regularity requirements: hence the behavior factor should be
suitably modified. To this purpose, it is necessary to study the specific features of the structure, in
terms of building material and geometric configuration
Chapter 6 of Eurocode 8 is entirely devoted to steel buildings. In section 6.3.1 the possible
structural types are briefly described:
a) Moment resisting frames, are those in·which the horizontal forces are mainly resisted by members
acting in an essentially flexural manner.
b) Frames with concentric bracings, are those in which the horizontal forces are mainly resisted by members
subjected to axial forces.

Appendix B
c) Frames with eccentric bracings, are those in which the horizontal forces are mainly resisted by axially
loaded members, but where the eccentricity of the layout is such that energy can be dissipated in seismic
links by means of either cyclic bending or cyclic shear.
d) Inverted pendulum structures, are defined in 5.1.2, and are structures in which dissipative zones are
located at the bases of columns.
e) Structures with concrete cores or concrete walls, are those in which horizontal forces are mainly resisted
by these cores or walls.
f) Moment resisting frames combined with concentric bracings.
g) Moment resisting franles combined with infills.
The frame studied herein can be attributed to the category a (moment resisting frames) with
respect to the transverse direction and b (frames with concentric bracing elements) for the longi-
tudinal direction where there are V-brace concentric elements; Figure B.5 shows the representation
of the categories a) and b), as reported in the Eurocode.
column). Default value for αu /α1 (see 6.3.2(3) and Table 6.2)
Figure B.5 - Moment resisting frames (dissipative zone in beam and at bottom of
Figure B.6 - Schematic drawing of moment resisting frames (which are

representative of the transverse frames in the present case) and frames with
concentric bracings (V-bracings are present in the longitudinal frames).

Appendix B
Section 6.3.2 of the Eurocode lists the requirements for the selection of the behavior factor:
“The behaviour factor q, introduced in 3.2.2.5, accounts for the energy dissipation capacity of the structure.
For regular structural systems, the behaviour factor q should be taken with upper limits to the reference
values which are given in Table 6.2 (Table B.6, editor’s note), provided that the rules in 6.5 to 6.11 are met.
If the building is non-regular in elevation (see 4.2.3.3) the upper limit values of q listed in Table 6.2 (Table
B.6, editor’s note) should be reduced by 20 % (see 4.2.3.1(7) and Table 4. I).”
Ducility Class
STRUCTURAL TYPE
DCM DCH
a) Moment resisting frames 4 5 αu / α1
b) Frame with concentric bracings

Diagonal bracings 4 4
V-bracings 2 2,5
c) Frames with eccentric bracings 4 5 αu / α1
d) Inverted pendulum 2 2 αu / α1
e) Structures with concrete cores or concrete walls See section 5
f) Moment resisting frame with concentric bracings 4 4 αu / α1
g) Moment resisting frames with infills

Unconnected concrete or masonry infills,
in contact with the frame 2 2
Connected reinforced concrete infills See section 7
Infills isolatd from moment frame
(see moment frames) 4 5 αu / α1
Table B.6 - Upper limit of reference values of behaviour factors

for systems regular in elevation

Appendix B
By considering a medium ductility class (DCM), the value of q is equal to 4 and 2 in the transverse
and in the longitudinal direction, respectively (it is possible to assign different values for the two
different directions), in case the regularity conditions in elevation are fulfilled. Since, however,
this requirement is generally not met in a satisfactory manner for the case of modular pipe-racks,
it is necessary to introduce a 20% reduction. The final values of the behavior coefficients scale
down to 3.2 for the longitudinal direction and 1.6 for the longitudinal frames.
The coefficient in the longitudinal direction deserves specific considerations: the inhomogeneous
values of the behavior factor in the two directions may finally yield a non-optimal seismic design.
It is thus important to understand if it is possible to raise the value in the longitudinal frames,
without changing the bracing scheme which is characterised by structural simplicity and full com-
patibility with the service requirements. It must be noticed that the standard imposes a low
behavior factor in the case of V-brace (chevron), since it presupposes a static scheme like the one
shown in Figure B.7.
Under this hypothesis, the tensile force in one diagonal is limited by the carrying capacity under
axial compression of the other diagonal, which may suffer by elastic buckling. It is possible to en-
visage a different static scheme, shown in Figure B.8, where the flexural response of the beam is
involved so that the tensioned brace can reach the plastic strength (Npl, Rd), even if the other
diagonal has already exceeded the bearing capacity connected to buckling (Nb , Rd ≤ Npl , Rd).
Figure B.7 - Static scheme for the standard interpretation of V-bracing system
In this case, the structural behavior is similar to eccentric bracing, which entails the plastic response
of the beam and which is characterized by q = 4 (to be reduce until 3.2 for the non- regularity in
elevation). Of course, the increased behavior coefficient is realistic only if the beams are capable

Appendix B
of withstanding the unbalanced vertical force at the tip of the V-brace. The vertical action on the
beam is given by the difference between Npl, Rd and Nb, Rd, projected in the vertical direction
(see Figure B.9)
Figure B.8 - Alternative interpretation of the static scheme

for the V-bracing system
It is possible to predict a moderate increase of the beam dimension, as a consequence of the

flexural involvement in the seismic response. Nevertheless, the parametric studies have shown
that the achievement of a uniform behavior coefficient in the two directions is by far more
important, in order to obtain the optimal seismic design.
Figure B.9 - Computation of the unbalanced vertical force which entails an

additional bending moment on the beams
B.2.2 Operational details of seismic analyses
As already mentioned, the seismic analyses have been carried out for two different earthquake
intensities, in order to enlighten the critical issues which are connected to the two scenarios for

Appendix B
both the Ultimate Limit States and the Service Limit States.
The scenario of moderate earthquake is represented by a PGA of 0.2 g. The design spectrum for the
horizontal actions (which are the most important in the specific case) is defined in the Eurocode 8:
In the previous formulas: Sd ( T ) is the design spectrum; T is the vibration period; ag is the chosen
PGA; TB is the lower bound and TC is the upper bound for the constant branch of the acceleration
spectrum; TD is the starting point of the branch at constant displacement; S is the soil coefficient;
q is the behavior factor; defines the lower bound for the design spectrum (recommended value
β = 0.2). The values TB , TC , TD , S depend on the soil type and are indicated by the national regu-
lations. In the specific case, the following values have been adopted: S = 1.2; TB = 0.15 s; TC = 0.5
s; TD =2.0 s.
The design spectrum has been introduced in the computer code, as graphically explained in
Figure B.10. In the case of severe earthquake conditions, the only modification is referred to the
PGA value, which is changed into 0.4 g. The final design spectrum is simply multiplied times a
factor 2 with respect to the previous case.
Figure B.10 -
Screenshot of the
data input for the
definition of the
design spectrum in
the case of moderate
seismic intensity

Appendix B
In order to satisfy the basic requirements for the seismic assessment, it is necessary to check the
following limit states: ULS and DLS. The load combination is obtained by considering the seismic
actions associated to the dead loads Gk,j and a portion of the live loads Qk,i
The combination factor ΨE,i accounts for the probability of simultaneity for the maximum credible
earthquake and the live loads. In the present case, by considering the recommendations of
Eurocode 8 and Eurocode 0 [10], a suitable value is ΨE,i = 0.3. Moreover, the seismic actions in
the two horizontal directions must be combined as follows (vertical excitation can be neglected in
this case):
The structural analysis has been carried out by means of a modal superposition technique (in the
hypothesis of conventional linear dynamic behavior). This analysis consists of the following steps:
(i) determination of the vibration modes of the building (modal analysis); (ii) calculation of the
effects of the seismic action represented by the design response spectrum for each of the vibration
modes; (iii) combination of these effects. Attention should be paid to the participating mass and
either of the following conditions should be satisfied: all modes with participating mass greater
than 5% are included in the analysis; a number of modes is considered such that the total partici-
pating mass is greater than 90%. For the combination of the effects of the individual modes, a
complete quadratic combination is adopted:
Ej is the effect of mode number j and ρij the coefficient of correlation between mode i and mode
j, which can be computed with the following formula:

Appendix B
ξ is the viscous damping coefficient and βij is the ratio between the period of mode j and mode i.
The combined actions are used for safety assessment. ULS are “associated with collapse or with
other forms of structural failure which might endanger the safety of people”, according to Eurocode
8. The safety assessment is carried out by checking the structural strength (with respect to various
failure modes, e.g. sectional resistance, overall and local buckling, etc.) and by verifying the
fulfilment of some geometric constraints which are connected to the proper development of dis-
sipative zones. It is useful to remind concisely the critical issues for the problem at hand.
Transverse frames (moment resisting frames)
» Beams. Sectional strength for bending, with possible influence of the shear force; buckling
for compressive force and for flexural-torsional interaction; limitation of the axial load in the
zones where plastic hinges may occur.
» Columns. Sectional strength for axial force, shear force and bending moment, with the
suitable overstrength factor in order to achieve a “strong column-weak beam” layout; lateral
and lateral-torsional buckling.
Longitudinal frames (concentric bracings)
» Diagonal members. Non-dimensional slenderness less than or equal to 2.0; sectional

strength for axial force; limitation of the difference between the maximum and the minimum
overstrength
» Beams and columns. Sectional strength for axial force, shear force and bending moment,
with the suitable overstrength factor in order to achieve dissipation in the diagonal members
only; lateral and lateral-torsional buckling.
DLS are studied “by satisfying the deformation limits or other relevant limits” defined in Eurocode
8. For the specific case, only the interstorey drift should be considered under a seismic action
having a larger probability of occurrence than the design seismic action corresponding to the ULS.
It is worth noting that the displacements due to seismic actions are obtained by applying the afo-
rementioned design spectrum and by multiplying the computed displacements times the “displa-
cement behavior coefficient”, which in general is assumed to be the same as the standard
behaviour factor. In the specific case, it is advisable to consider the presence of brittle non-
structural elements, so that the following inequality must be satisfied:
dr ν ≤ 0,005h
where is the design interstorey drift and is the storey height. The coefficient is introduced in order
to take into account the lower return period of the seismic action; in the present case, it has been
assumed .

Appendix B
B.2.3 Safety assessments for the original layout
The results of safety assessments for ULS are presented first. Both seismic intensities lead to the
violation (with different levels) of the design requirements which are summarized in the next list.
1) Column strength in the transverse frame: some columns at the top storey do not fulfill the
strength requirement in terms of combination of axial forces and bending moment (Figure B.11).
Figure B.11 - Schematic view of the columns which do not satisfy the safety
requirement in terms of sectional strength (combination of axial force and bending
moment with suitable overstrength factor for the moment resisting frame in the
transverse direction)

Appendix B
2) Column strength in the longitudinal frames: in this case, the violation happens at the base level
and it is connected to the requirements of capacity design in the case of concentric V- bracing
system; the critical columns are graphically identified in Figure B.12.
Figure B.12 - Schematic view of the columns which do not satisfy the safety
requirement in terms of sectional strength (axial force with suitable overstrength
factor due to the presence of dissipative V-bracing systems)
3) Diagonal bracings (Figure B.13): in spite of a positive compliance with strength and buckling re-
quirements, it has been found a violation in the distribution of the overstrength coefficient: more
specifically, the ratio between the maximum and the minimum overstrength value is 3.98 to be
compared with the upper limit Passing to DLS, it is possible to verify a complete fulfillment of the
requirements in the case of moderate earthquake. Conversely, for severe seismic actions, the in-
terstorey drift is excessive for the base storey in the transverse direction (Figure B.14).

Appendix B
Figure B.13 - Schematic view of the diagonal elements which do not satisfy the
safety requirement in terms of overstrength homogeneity.
Figure B.14 - Representation of

the deformed configuration in
case of severe earthquake: it is
evident that the maximum
interstorey drift is located at the
bottom storey.

Appendix B
B.2.4 Design suggestions for increasing the seismic performance
The violation of certain seismic requirements suggest the introduction of structural modifications,
some of which have been already presented in the analysis of wind action. In particular, the intro-
duction of a horizontal bracing system is beneficial for the collabouration of the transverse
frame, and it is also useful in the seismic case because it involves a certain degree of “floor action”
(despite the impossibility of introducing rigid floors at each storey level). The replacement of I
shaped columns with concrete filled hollow columns is a good provision in order to increase
the sectional strength: it is possible to verify that such a modification is sufficient to obtain the ful-
filment of safety requirements for ULS with a good margin. Moreover, the composite columns are
endowed with higher stiffness, so that it is possible to envisage an improvement of the safety as-
sessment for DLS. Nevertheless, the stiffness increase is not enough in this case. In fact, account
taken of the storey height = 4.6 m, one finds a limit threshold equal to 23.4 mm. Unfortunately,
the seismic analysis yields an interstorey drift equal to 26.08 mm, which means a modest violation
of the upper threshold.
A possible solution for the problem of excessive interstorey drift can be found by increasing the
cross section of the transverse beam. The minimum section required is, in the present case, is
HE500B, which entails a maximum interstorey drift equal to 22.88 mm. This provision is connected
with a conspicuous increase of the structural weight: the original layout included HE360A, with
mass per unit length equal to 112 kg/m; the new beams HE500B weigh 187 kg/m, with an
increment of 67%. Additionally, the change of cross section is against the concept of standardization:
view that the DLS violation happens only for severe earthquake, it would be desirable to keep the
basic features of the standard design that is appropriate for most cases (i.e. moderate earthquake
or lower). These considerations lead to the choice of a different stiffening strategy, based on the
addition of structural element rather than on the replacement of beams. A possible solution is re-
presented by transverse bracing, which however should not be so light if they had to satisfy
the strength requirements for ULS [11]. In order to obtain an optimal solution in terms of structural
weight, it is conceivable the introduction of a bracing system which is just used for displacement
limitation, with a limited (or, in the best case, absent) effect on the ULS. Such an objective might
be reached by the introduction of suitable mechanical fuses, for instance at the level of bracing
joints, which exclude completely the diagonal elements in the case of horizontal actions higher
than a certain threshold. However, it must be considered that this solution is connected with
several difficulties, not least the economic impact of joint replacement after seismic events. It is
better to envisage a bracing system where the compressed diagonal is expected to buckle under
ULS actions, so that the most restrictive safety requirements can be circumvented and a lighter so-
lution can be obtained. As a matter of fact, has it happens for the already discussed case of longi-
tudinal bracing, this situation implies an additional bending action on the beam, which should be
properly tackled.
The proposed transverse bracing is depicted in Figure B.15: the particular layout has been chosen

Appendix B
in order to minimize the interference with the pipes and other plant devices, which run mainly in
the longitudinal direction.
The achievement of a positive displacement assessment require really slender element, for instance
a couple of UPN80 paired along the longer side at a distance of 25 mm. In this way, the interstorey
drift at the bottom level is drastically reduced to 8.62 mm, thus respecting largely the safety limit.
The relative displacement at the upper storeys are not changed with respect to the original layout,
so that one can obtain a positive DLS assessment. Finally, it is necessary to check the ULS resistance
of the beam, which is subject to the unbalanced vertical action as shown in Figure B.15: it is
possible to verify that the increment of bending moment is considerable, though not sufficient to
run out the plastic strength of the original beam. No change of the beam’s cross- section is
required, and the basic concept of standardization is preserved.
Figure B.15 - Proposed modification for the improvement of DLS performance:

transverse bracing system.

Politecnico di Milano is a scientific-technological university which trains engineers, architects and
industrial designers. The University has always focused on the quality and innovation of its teaching
and research developing a fruitful relationship with business and productive world by means of
experimental research and technological transfer.
Research has always been linked to didactics and is a priority commitment which has allowed Po-
litecnico di Milano to achieve high quality results at an international level as to join the university
to the business world.
Research activity moreover constitutes a parallel path to that formed by cooperation and alliances
with the industrial system. The alliance with the industrial world, in many cases also favored by
the Fondazione Politecnico and by Consortiums to which Politecnico belongs, allows the university
to follow the vocation of the territories in which it operates and to be a stimulus for their develop-
ment. The challenge which is being met today projects this tradition which is strongly rooted in
the territory beyond the borders of the country, in a relationship which is developing first of all at
the European level with the objective of contributing to the creation of a "single professional
training market".
Politecnico takes part in several research and training projects collaborating with the most qualified
European universities. Politecnico's contribution is increasingly being extended to other countries:
from North America to Southeast Asia to Eastern Europe. Today the drive to internationalization
sees Politecnico di Milano take part in the European and world network of leading technical uni-
versities and offers several exchange and double-degree programmes beside 2 three-year Bachelor
of Science degree programmes, 10 Master of Science degree programmes, 12 Specialization
Master's programmes and 24 Doctoral programmes which are entirely taught in English
www.polimi.it

ANIMP is the Italian Association of Industrial Plant. Established in 1974, ANIMP is the “meeting
centre” where Engineering Companies, Manufacturers of Plant Components and Universities may
develop innovations and share common experiences to foster the construction of reliable and
efficient industrial plants.
The joint goal of all members of ANIMP is to conceive and implement those improvements and
changes that will enable them to respond quickly, with the right skills and technology to the chal-
lenges that world-wide Customers demand from industrial plants.
ANIMP Missions
» To represent and promote Italian plant industry (engineering, engineering and contracting
companies, manufacturers of components for plants)
» To carry out events to increase the cultural and scientific knowledge, necessary for ANIMP
Members to compete in the international market
» Indeed, through its Members, ANIMP is a highly qualified and prestigious vehicle of “Sistema
Italia” in the world of the Plant Engineering & Supply Industry.
» ANIMP Members rely on the vast resource of more 95.000 highly skilled engineers and te-
chnicians.
www.animp.it

Task force (members)
Luigi Anselmi
Head of Construction Operations
Tecnimont (Maire Tecnimont Group)
Raffaele Ardito
Assistant Professor - Department of Civil and Environmental Engineering
Politecnico di Milano
Angelo Balestrero
Senior Construction Manager
Nicola Careri
Research fellow - Dept of Management, Economics & Industrial Engineering
Paolo Cremonini
Director of Operations
Fagioli
Marco Di Gennaro
Support Services Section Head
Technip Italy
Luca Ferrazzi
Construction Methods Department Manager
Saipem
Gabriele Franco
Construction Director B.U. Oil & Gas
Rosetti Marino
Alistar Gibb
Professor
ECI Royal Academy of Engineering

Mauro Mancini
Assistant Professor, ANIMP Construction Section Coordinator
Dept of Management, Economics & Industrial Engineering - Politecnico di Milano
Andrea Manuli
Lead Civil and Structural Engineering
Technip Italy
Pierfranco Mocchetti
Civil & Structural Technical Manager
Alessandro Palmeri
Senior Lecturer in Structural Engineering - School of Civil and Building Engineering
Loughborough University
Federico Perotti
Full professor - Department of Civil and Environmental Engineering
Marco Spinelli
Head of Project Engineering and Technical Support Services
Paolo Trucco
Associate professor - Dept of Management, Economics & Industrial Engineering
Flavio Vitalini
Civil Engineer Manager
Foster Wheeler
Angelo Zucconi
Senior Project Construction Manager
Serding

Edited by Mauro Mancini
Graphic design: Paolo Sicignano (www.sicignano.net)
Print: Tipografia Litografia A. Scotti Srl
Via E. Berlinguer n. 6
20872 Cornate d'Adda (MB)
Source:
cover photo by Saipem
Man, Woman by Edward Boatman from The Noun Project

Companies involved

White Book Mancini

Uploaded by

Copyright:

Available Formats

White Book Mancini

Uploaded by

Document Information

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

White Book Mancini

Uploaded by

Copyright:

Available Formats

What is the purpose of the publication?

What is the purpose of the publication?

What is ANIMP's main mission?

What is ANIMP's main mission?

Advances in plant

Edited by Mauro Mancini

© ANIMP SERVIZI srl

with the support of

Advances in plant modularisation / 3

4 / Advances in plant modularisation

Advances in plant modularisation / 5

6 / Advances in plant modularisation

» a systematic comparison between concepts expressed in literature and actual practices;

» an analysis of the identiﬁed technical criticalities focused on providing general and on

Advances in plant modularisation / 7

8 / Advances in plant modularisation

» Very high labour cost at site

» Very low labour productivity at site

» Wide lack of skilled manpower in the region/area of the site

» Lack of adequate infrastructure to host a high number of expat workers

» Restrictions on maximum number of allowed expat workers

» Constraints on maximum number of workers simultaneously operating at site

» Serious safety/security concerns at site

» Need for a project crashing

Advances in plant modularisation / 9

» Environmental, legal or regulatory constraints at site

Improved quality Higher Security

Social/Environmental Impacts reduction Lower Manpower costs

Reduction of weather impacts

Higher transportation costs Higher need for infrastructure

Local content impacts ...

Table 1.2 - Modularisation drawbacks generally recognized by literature

10 / Advances in plant modularisation

Table 1.3 - Table of common terms

1.2 Survey features

Advances in plant modularisation / 11

» Deﬁnition, application areas and pro&con stakeholders

» Driving factors and objectives

Table 1.4 - Survey summary

» plant decomposition according to both constructability and process logic

12 / Advances in plant modularisation

» prefabrication of the modules in yards other than the construction site

» a need to satisfy transport and assembly requirements

AREAS OF INTEREST MEAN

Figure 1.3 - Average weighted modularisation areas of interest

Advances in plant modularisation / 13

Figure 1.4 - Installation at site of a module fabricated by Saipem at Arbatax yard

1.3.2 Driving factors and objectives

14 / Advances in plant modularisation

Figure 1.5 - Average weighted modularisation drivers

Advances in plant modularisation / 15

16 / Advances in plant modularisation

Advances in plant modularisation / 17

Increasing construction efficiency

Reducing project delivering time

Reducing project cost

Enhancing market competiveness (e.g. better services)

Enhancing project manageability

Balance standardization with customization