Professional Documents
Culture Documents
Alex C. Arthur
This edition first published 2022
© 2022 John Wiley & Sons Ltd
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
except as permitted by law. Advice on how to obtain permission to reuse material from this title is available
at http://www.wiley.com/go/permissions.
The right of Alex C. Arthur to be identified as the author of this work has been asserted in accordance
with law.
Registered Offices
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
9600 Garsington Road, Oxford, OX4 2DQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products
visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that
appears in standard print versions of this book may not be available in other formats.
10 9 8 7 6 5 4 3 2 1
To my dearest mum, Mama Alice Felicia Arthur, and my beloved son,
Alex Collins Arthur Jnr (aka Xander)
Mum, the inspirational words that you said to me at the tender age of 10 have stayed with me
and given me the confidence to pursue excellence. I will always treasure those words.
Xander, you have given me a sense of accomplishment since you came into my life.
vii
Contents
Preface xi
Acknowledgement xiii
About the Author xv
Part 1 Concepts 5
7 Data Presentation 89
7.1 Introduction 89
7.2 Case Study Project 1 89
7.2.1 Case Study 1 Participants 90
7.2.2 Case Study 1 Findings 91
7.2.2.1 Case 1 – Research Proposition 1: Findings 91
7.2.2.2 Case 1 – Research Proposition 2: Findings 102
7.2.2.3 Case 1 – Research Proposition 3: Findings 124
7.2.2.4 Summary of Case 1 Findings 146
7.3 Case Study Project 2 147
7.3.1 Case Study 2 Participants 147
7.3.2 Case Study 2 Findings 147
7.4 Case Study Project 3 150
7.4.1 Case Study 3 Participants 151
7.4.2 Case Study 3 Findings 153
7.5 Case Study Project 4 154
7.5.1 Case Study 4 Participants 154
7.5.2 Case Study 4 Findings 155
7.6 Summary 158
x Contents
8 Application 161
8.1 Introduction 161
8.2 Research Proposition 1: Discussions 161
8.2.1 Risk Perception Categorisation on the Typical Construction Project Risk Events
at the Different Project Delivery Phases 162
8.2.1.1 Pre-construction Phase 162
8.2.1.2 Construction Phase 163
8.2.2 Risk Perception Categorisation on the Typical Construction Project Risk Events
Under Different Project Settings 164
8.3 Research Proposition 2: Discussions 169
8.3.1 Intuitive Risk Management Decision Processing from ‘Grounded’
Heuristics 169
8.3.2 Susceptibility of Intuitive Decision Processing to Manipulation 173
8.3.3 Psychological Difficulties in Intuitive Risk Identification of Events Outside the
Scope of a Specialist Role 173
8.4 Research Proposition 3: Findings 179
8.4.1 Two Strands of Intuitive Construction Risk Management Systems 180
8.4.2 Theoretically Incompatible Risk Management Practices 184
8.4.3 Intuitive Processing of Statistics and Probability Data 186
8.4.4 Comparative Analysis of Intuitive Processing of Quantitative Risk Assessment
Versus Qualitative Risk Assessment 189
8.4.5 Intuitive Processing of Probability Predictions of Emotive Events 192
8.5 Summary 196
8.5.1 Research Proposition 1 196
8.5.2 Research Proposition 2 197
8.5.3 Research Proposition 3 198
9 Conclusions 201
9.1 Summary 201
9.1.1 Research Proposition 1 202
9.1.2 Research Proposition 2 203
9.1.3 Research Proposition 3 204
9.2 Rethinking Construction Risk Management Practices 205
References 391
Index 401
xi
Preface
There are extensive psychometric research and publications that confirm the influence of
behavioural stimulus in intuitive risk management decision-making systems (Kahneman
and Tversky 1982a; Tversky and Kahneman 2002; Finucane et al. 2003; Slovic 2010;
Kahneman 2011). The existing construction management publications however appear
limited in exploring the behavioural patterns underpinning the high incidence of intuitive
construction risk management practices. The purpose of this book therefore is to address
the data gap, by extending current empirical and analytical evidence from systems thinking
and behavioural sciences into construction risk management research, to strengthen their
theoretical basis and further illuminate emerging theoretical extensions.
The empirical findings from four case qualitative research methodology have revealed
substantial evidence in support of three theoretical propositions: risk perception categorisa-
tion within the construction project delivery system reflecting the structure of the differen-
tiated specialist roles, psychological difficulties associated with intuitive risk identification
of events outside the scope of our heuristics, and incompatibilities of mixing decision pro-
cessing approaches and data presentation formats from different systems of thinking and
decision-making.
The subsequent analytical discussions have highlighted the need to rethink construction
risk management practices, by introducing behavioural science training to construction
project management students and professionals. This will equip them with the requisite
competencies to identify and harmonise the different affective heuristics, as well as the
technical variations of the specialist roles involved in project delivery. There is also the
need to consider revising construction risk data presentation from statistics and probability
to qualitative formats to complement the current intuitive construction risk management
decision-making practices.
xiii
Acknowledgement
I am thankful to God for granting me wisdom and insight throughout the book project.
I would also like to pay tribute to my former doctoral supervisor, the late Professor
Emeritus Stephen D. Pryke, for his tremendous influence on the development of my initial
research ideas. I must also mention the support received from the lecturers and staff of
the Bartlett School of Construction and Project Management, University College London,
especially Professor Hedley Smyth.
Furthermore, I wish to express my sincere gratitude to my dear parents, Mr and Mrs
Arthur, my wife, Mrs Rubina F. Arthur, my beloved son, Alex Collins Arthur Jr, and all my
siblings for their love and support.
Lastly, I would like to thank the directors and staff of the organisations that participated
in the data collection processes, for the Part 2 case studies.
xv
Dr Alex C. Arthur, BSc MSc PhD FRICS MAPM is a chartered project management
surveyor with more than 14 years’ experience working in multidisciplinary practices. He is
a Fellow of the Royal Institution of Chartered Surveyors (RICS) and an examiner of their
Assessment of Professional Competency (APC) programme. He is also a member of the
Association of Project Managers.
Dr Arthur is a certified professional teacher and postdoctoral researcher who possesses
extensive research and industrial expertise within the United Kingdom and other interna-
tional construction industries. He has published a number of articles, journals, and con-
tributions to books on systems thinking and analysis, intuitive decision-making, intuitive
construction risk management systems, construction risk management decision-making
systems, partnering procurement, development control, and planning systems.
1
Hazard and uncertainty have always been a part of human condition (Lock 2003; Beck
2007). The early humans at pre-civilization faced the daily challenge of protecting them-
selves against attacks from wild animals and hostile tribes. The modern society may have
developed remedies against those threats, but there are equally other factors that threaten
our persistence, like the ongoing Covid-19 pandemic. The emergence of vaccines has pro-
vided glimmers of hope that we may be heading towards the end of the pandemic. The fact
however remains that there will always be other natural and social menace that challenges
human survival.
Social hazards and uncertainties were conventionally defined and managed by the reli-
gious and political leaders occupying the top of the vertical social structure. This was done
through the reliance on guided morality, superstition, taboos, and rituals (Japp and Kusche
2008). The contemporary transition of multiple differentiated functional subsystems replac-
ing the stratified vertical system has expanded social risk communication into the public
realm (Japp and Kusche 2008). According to Loosemore et al. (2006), the increased public
awareness on the risk variables associated with both personal and corporate activities and
increased media reportage on the impact of risk events have accounted for the heighten
discussions on risk and influenced public attitude to risk management. A failed suicide
bomb attack in which the perpetuators planned to detonate liquid explosives on several
aircrafts traveling from Heathrow Airport to the United States of America in 2006 led to
public discussions on the risk posed by liquids on aircrafts, resulting in a European Union
imposition of restrictions on carrying liquids on aircrafts, which subsequently has become
a world-wide ban. The use of Blackberry Instant Messenger system to elude police intelli-
gence by the perpetuators of the August 2011 British riots triggered public discussions on
the security risk associated with smart phones (The Economic Times 2011).
In the context of construction delivery, the catastrophic demise of a 6-year-old girl at Car-
nival Place, Moss Side, Manchester, United Kingdom on 28 June 2010, and a 5-year-old girl
at Brook Court, Bridgend, South Wales, United Kingdom on 3 July 2010, after they became
trapped in electric gates, coupled with two other near misses in July and September 2010,
ignited public debates on the safety of electric door entry systems. The result was the for-
mation of the Gate Safe charity in 2010, to promote safety guidance for the construction
of automatic and manual gates (Gate Safe 2020). The 2017 Grenfell Tower fire incidence
in the United Kingdom has also stimulated public discussion on the fire safety risks asso-
ciated with high rise buildings (Arthur 2017). The immediate impact has been a public
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
2 1 Introduction – A Risk Management Approach to Construction Project Delivery
inquiry and enactment of changes to the statutory building regulations. The heightened
public risk communications have cumulatively increased construction clients’ awareness
of the typical industry risk events, with the expectation of their management through the
project delivery processes. The control of project risk is now seen as being synonymous with
the control of the project itself (Pryke and Smyth 2006; Abderisak and Lindahl 2015), which
in effect establishes a direct relationship between effective risk management systems and
project success (Cagno et al. 2007).
Risk as explained by Loosemore et al. (2006) is an uncertain event that might occur in the
future, to potentially affect an interest or objective (usually adversely), although the precise
likelihood or impact may be indeterminate. The theoretical emphasis is the relationship
between risk and uncertainty and probability judgement. It also emphases risk as a future
potential occurrence with the propensity to impact on an objective or interest.
Loosemore (2006) has also described risk management as a field of competing ideologies
between the homeostatic and callibrationist perspectives (Hood and Jones 1996). The home-
ostatic perspective proposes scientific approach to risk management through structured risk
identification and rational decision-making, whilst the callibrationist viewpoint believes
in human subjectivity in risk interpretation, identification, and treatment (Loosemore
2006). Another significant ideology of the callibrationist view is the subjectivity and biases
of the personal perceptions responsible for guiding the risk identification and treatment
processes (Tversky and Kahneman 1982a,b,c; Kahneman and Tversky 1982a,b; Slovic 2010;
Kahneman 2011). Japp and Kusche (2008) theorisation of social construction of risk
through the communicative and decision-making processes of multiple differentiated
functional subsystems and the ensuing analytical variability in social risk interpretations
give the callibrationist approach an edge over the homeostatic method in the management
of risk within the contemporary era.
This book has therefore adopted theories and concepts from systems thinking and
behaviour sciences in demonstrating the social construction of risk (Zinn 2008) and the
subjectivity of risk perceptions (Slovic 2010; Kahneman 2011). The principles of general
systems theory (Bertalanffy 1968, 2015) have enabled conceptual analysis of the risk
management subsystem within the construction project delivery system. The concept
of systems differentiation (Walker 2015) and the existing empirical evidence on the
psychology of perception (Slovic 2010; Kahneman 2011) have been applied in analysing
the pattern of risk perceptions emanating from the different specialist roles responsible
for design development and risk management decision-making, including the project
manager, contracts manager, technical manager, architect, engineer, quantity surveyor,
and client. The emphasis on design development is due to the close relationship between
design concepts and the generation of project risk events (Latham 1994; Egan 1998; Lock
2003; Flyvbjerg et al. 2003). According to Bea (1994), the bulk of the initial project design
errors later results in risk events, with approximately 42% and 50% actualising during the
construction and operational phases, respectively.
The ensuing analytical review has resulted in the discovery of a new theoretical interpre-
tation for risk perception categorisation based on the differences in affective heuristics of
the different specialist roles within the construction project delivery system. The theoreti-
cal evidence has been further confirmed through a four case qualitative empirical research
investigation covered in the Part 2. There has been further theoretical discussion on the risk
1.1 Risk Perception Categorisation 3
The theoretical evidence on systems differentiation (Walker 2015) and systems decom-
position processes (Carmichael 2006) confirms the presence of micro differences in
objectives within systems components. This in turn suggests differences in how the
internal parts experience and respond to the impact from their environmental forces
(Loosemore et al. 2006). The implications being that mixing tools and techniques from
different decision-making systems, as evident in the present construction risk management
practice of intuitive processing of statistical and probability data (Akintoye and Macleod
1997; Lyons and Skitmore 2004; Kululanga and Kuotcha 2010; Bowden et al. 2001; Lock
2003) may introduce conflicting responses which may impede the required collaborative
transformational processes, thereby producing systemic errors (Kahneman 2011). The
ensuing theoretical issues and psychological concerns have been analysed in the Part 1 and
further validated through the case study empirical investigations presented in the Part 2.
5
Part 1
Concepts
7
Figure 1 provides an overview of the Part 1 (Chapters 2–5), which discusses the key concepts
and theories covered in the book.
Chapter 2
Takes a social construction risk approach through the application of systems thinking
and analysis to model the construction industry: structure, internal project delivery
and management processes, and interactions with the wider environment, leading to
risk generation.
Chapter 3
Discusses the concept of risk, including risk etymology, and the different conceptual
interpretations of risk. Concludes with an extensive review of the sociological and
psychological risk perspectives applied in the book.
Chapter 4
Chapter 5
Applies the psychometric risk conception to discuss the different systems of thinking
and decision making, the psychology of perceptions including risk perception
formation; impact of affect heuristics on cognitive reasoning; construction risk data
presentation formats and affective heuristics. Further discussions on intuitive
decision processing including the theoretical issues associated with the intuitive
processing of statistical and probability data. Concludes with a review of the
prevailing theories on risk perception categorisation, leading to the proposition of a
novel theory, the differentiated risk perceptions
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
9
2.1 Introduction
Chapter 2 begins the conceptual review. Taking a social construction approach to risk inter-
pretation, the principles of systems thinking and analysis are applied in discussing the con-
struction industry; internal structure, transformational processes, and the interactions with
the wider environment, leading to risk generation. The chapter starts with an exposition of
the UK construction industry, its contribution to the national economy and government’s
aspiration to reduce project capital cost, whole life costing, and construction period. The
government has also set targets for reduction of greenhouse gas emission in the built envi-
ronment, and the trade gap between export and importation of construction products and
materials. The high standard industry aspirations call for efficiencies in the project man-
agement and operational systems, including risk management practices.
The chapter then proceeds to review some key principles of general system theory includ-
ing open systems whose operations involve interactions with the other systems and subsys-
tems within the wider industry environment, and closed systems which are self-containing
entities, able to function independent of external influences. The constant interactions of
the human and material resources within the construction industry accordingly confirms
its classification as an open system. There is also theoretical review of systems objectives
which defines the required inputs, transformational processes, and outputs. The subse-
quent analytical review of the internal structure of the construction system, using the con-
cept of systems decomposition, reveals the core components comprising of different special-
ist roles, department, organisations, and sectors providing the human resources; different
operational phases and stages for the product delivery; and different tools and techniques
applied in the conversion of construction materials and resources into the final products.
The theoretical review continues with the examination of the construction delivery sys-
tem through the RIBA Plan of Work 2020 stages. The input of a stage is generated from the
output of the preceding stage and the wider construction industry environment. The trans-
formational processes apply construction industry tools and techniques to convert received
inputs into expected outputs, for the subsequent delivery stages and the wider construction
industry environment.
The construction delivery system involves constant interactions of different specialist
roles with different functionalities. The failure to carefully coordinate the multifaceted
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
10 2 Systems Analysis of the Construction Industry and Project Delivery
human needs. There are also statutory and operational policies that regulate construction
activities. The construction delivery process goes through different stages where human
and material resources are transformed into finished construction products. This has led to
some construction management researchers such as Walker (2007, 2015) arguing similari-
ties between the construction delivery process and the principles of systems thinking and
analysis.
A system is a set of related variables which function together as a group, and in the
process, displays the prime attributes of the collective group instead of the individual com-
ponents (Checkland 1999). Systems thinking and analysis on the other hand is an inves-
tigative approach to studying a subject by adopting the principles of classical natural sci-
ences (Checkland 1999). The underlining principles originated from the biological sciences,
through the discovery of the general systems theory in the 1960s (Walker 2007, 2015; Berta-
lanffy 1968, 2015). This occurred at a time when Ludwig Von Bertalanffy was investigating
the linkages between the different science disciplines, which hitherto were studied as dis-
tinct subjects. Bertalanffy has suggested that the application of the principles of systems
thinking pre-dates the discovery of the general systems theory, and may have been applied
in earlier studies including Paracelsus’ research in medicine, and Leibniz’s studies in philos-
ophy, although the term ‘system’ may not have been specifically mentioned. Arthur (2020)
has also argued the application of systems analysis in the Christian Holy Bible narratives,
as seen in Apostle Paul’s comparison of the Christian mission to the corporate functioning
of the different human body parts (Romans 12:4–8; 1 Corinthians 12:12–31). The ideolo-
gies of systems thinking and analysis are not only limited to the classical natural sciences
but could also be applied in other disciplines to investigate the networks and interactions
between the sub-modules. Checkland (1999) has therefore suggested that systems thinking
and analysis should be interpreted as a methodology for studying a subject, rather than as
a distinct discipline.
System environment
Figure 2.1 Open system model. Sources: Adapted from Cole (2004), Cole and Kelly (2020), and
Walker (2015).
An open system on the other hand is active and adaptive to changes within its environ-
ment (Cole 2004; Cole and Kelly 2020; Wetherbe and Vitalari 1994). Cole and Kelly (2020)
has suggested the key characteristics of an open system as receiving inputs or energy from its
surrounding, converting the inputs into outputs, and subsequently discharging the output
back into its environment.
An open system’s analytical approach to studying a subject incorporates both the inter-
actions within the system’s internal structure and processes and the networks and link-
ages within the wider environment (Walker 2015). The networks and interactions result in
the conversion of inputs from the environment through the system’s internal transforma-
tion processes into outputs to serve the other systems within the wider environment (see
Figure 2.1).
In practice, living subjects exhibit open system characteristics by taking their energy from
their environment and using their internal transformational processes to convert into out-
puts to serve the other systems within their environment (Mullins 2005). An example of an
open system is automobile vehicles that receive inputs in the form of fuel, lubricants, and
coolant from the petroleum and vehicle consumable industries, and then converts through
their internal mechanisms into energy to propel the vehicles in transporting human and
other resources from the other systems within its environment. Likewise, the construc-
tion industry as an open system receives inputs in the form of specialist skills and material
resources from its environment, and then converts through the application of industry man-
agerial and operational principles, into construction outputs such as residential dwellings,
offices, industries, roads, bridges, hospitals, schools, commercial outlets, and airports, for
the benefit of the other systems within the wider environment. Wetherbe and Vitalari (1994)
have argued that few systems in reality behave exclusively as opened or closed. Most systems
function in a continuum between the two extremes.
Construction systems’ objectives similarly are responsible for directing the integrated
human and material resources from project inception, through the different transforma-
tional phases, to the desired outputs. Cole (2004) has suggested differences in construction
systems objectives at the strategic and operational levels. The strategic objectives define
the broad overall organisational mission statements and long-term business strategic aims,
whilst the operational objectives outline the individual project expected output (Carmichael
2006).
Construction systems’ objectives also function as parameters for monitoring and
appraising performance at both the strategic and operational levels (Lock 2003). Appraisal
at the operational level occurs at the interim delivery phases and overall completion
stage. Project execution outputs are constantly evaluated against the contemporaneous
project objectives at the different delivery phases, before progressing to the next phase.
When a project achieves completion, the final product is assessed against the project
brief to establish success or failure. The effectiveness of a construction system’s objective
in assessing project performance largely depend on it being clearly defined, to avoid
ambiguity in interpretation and application.
Walker (2015) has argued that, construction project objectives are generated as a
response to demands from client organisational stakeholders, comprising customers,
employees, shareholders, suppliers, communities, and related statutory authorities. Cova
and Salle (2006) have also classified organisational stakeholder groups into business
actors and non-business actors. The business actors consist of consultants, agents, project
financers, sub-consultants, and sub-contractors. The non-business actors are also made up
of government organisations regulating construction activities, construction trade unions
and pressure groups, syndicates, lobbies, and activists.
The analysis of stakeholder demands within the framework of client organisational busi-
ness objectives usually results in responses ranging from construction of a new project,
refurbishment or extension of an existing construction asset, outsourcing additional build-
ing space through direct purchase, leasing, or renting (Winch 2010). This applies to both
clients whose core business activities involves the provision of construction products and
also clients who require construction products to support their core business activities. As
an illustration, a residential property developer will most likely commence a new develop-
ment when there is evidence of demand for housing within its areas of business activities,
either in the form of increased applicants on its waiting list or improved economic activities
which draw in additional potential customers. In the same way, a supermarket will extend
or build a bigger shop when there is increase in its customer base.
According to Carmichael (2006), open systems exist within their wider environment as
subsystems interacting with allied subsystems to define the boundaries of their superior
main system. The multiple related open systems subsequently become the components
of the superior main system. Applying the principles of systems thinking correspondingly
enables analysis of the components of the subsystems, as sub-subsystems with parts inter-
acting to define the boundaries of the respective subsystems. The micro parts can be further
analysed as sub-sub-subsystems also exhibiting general systems properties. The process
could be continued in a linear progression to give an analysis of systems decomposition.
The application of systems decomposition at the construction industry organisational
level as depicted in Figure 2.2 identifies the different client, contractor, and consultant
organisations, which collectively forms the construction industry’s system. Within the three
broad classifications are subsystems which could be categorised based on their geographical
locations, scope of core business activities, type of business ownership, level of experience,
and type of services and products. Narrowing it down to the individual companies and firms
are devolved organisational structures in the form of departments and teams which could be
analysed as sub-subsystems. The departments and teams further devolve into lower organ-
isational structures of programmes and projects. The multiplicity of categorisations within
the client, professional consultancy, and building contractor organisations and the ensuing
lower tier micro systems, coupled with the varied interactions existing between the compo-
nents, suggest the construction systems decomposition analysis could be complex than the
simplistic model depicted in Figure 2.2.
Systems decomposition at the construction project organisational level can also be anal-
ysed through the categorisation of project objectives as subsystems. The various transfor-
mational processes and tasks responsible for the achievement of the different objectives
become sub-subsystems. The required activities within the various transformational pro-
cesses and tasks subsequently become the sub-sub-subsystem.
Professional Building
consultancy contractors’
firms subsystem
subsystem
Residential sub-subsystem Commercial/
retail sub-
Firm A Firm B subsystem
Firm C
Road and
bridges sub-
Leisure sub- subsystem
subsystem
Firm A
Department
B
Department A
P2
P1
PP2
PP1
P3 Department
C
P = Programmes
PP = Projects
The second stage of the RIBA Plan of Works centres on the preparation of project brief.
This involves feasibility studies of the favourable development proposal established in
the preceding stage, using site survey results and existing documentation. The client may
appoint a design team comprising of architectural and engineering firms to develop design
proposals. The stage output in the form of a project brief outlines the key project outputs
including design quality aspirations, agreed budget, project delivery programme, and
project execution plan.
The next stage is the development of architectural design concepts based on the project
brief developed at the second stage. The project design team comprising of architects and
engineers produces architectural design concepts integrating strategic engineering and
statutory design requirements. The design concepts must also align with the project budget
and the strategic quality and spatial aspirations established as part of the project brief.
Subsequently, the design concepts are reviewed with the client and project stakeholders
to establish and agree the design elements that deviates from the project brief. The stage
outputs include agreed architectural design concepts and schedule of derogations, agreed
revised project delivery programme and budget.
The fourth stage involves the development of architectural and engineering informa-
tion including spatial coordination. The key activities at this stage are design development
and reviews including engineering analysis, with corresponding project cost evaluations to
result in spatially coordinated designs which meets the project budget, quality standards,
and delivery programme. Planning application for the delivery of the project is usually sub-
mitted at this stage. Also depending on the chosen procurement route, the client may enter
into a pre-construction services agreement with a building contractor, where a two-stage
procurement approach is applied.
The fifth stage termed technical design, involves further development of the spatial coor-
dinated designs for the construction of the project. The design development at this stage
comprises architectural and engineering technical designs and integration of specialist
building services design information. Project procurement in the form of preparation and
issue of invitation for tender to appoint a building contractor is undertaken at this stage.
There is also submission of design information for building regulations application, dis-
charge of reserved pre-commencement planning conditions, preparation of construction
phase plan, and submission of F-10 to the Health and Safety Executive, in line with the
requirements of the Construction (Design and Management) Regulations 2015.
The next stage termed manufacturing and construction is where the actual site construc-
tion activities take place. The core tasks at this stage include finalisation of site logistics
arrangement, manufacturing and construction of the building system, monitoring of the
site works against the project quality standards and programme, installation and commis-
sion of specialist building systems, compliance and discharge of reserved building regu-
lations and planning conditions, compliance of the Construction (Design and Manage-
ment) Regulations 2015 including the maintenance of appropriate construction health and
safety plan, collation of operational and maintenance manual for submission at the project
handover.
The seventh stage is when the site construction works are completed, and the project is
handed over to the client. The handover activities include inspection and rectification of
2.5 The Construction Project Management System; Differentiation and Risk 17
RIBA Plan
of Work 2020 Input Transformation Output
Project management Planning: What tasks are involved and in what Residential
tools and techniques order? What materials and labour inputs are
accommodation
required?
Feedback
Figure 2.3 The construction project management systems. Sources: Adapted from Cole (2004),
Carmichael (2006), Cole and Kelly (2020), and Walker (2015).
The ensuing multiple divergent responses if left unresolved, usually leads to cross-purpose
working practices and conflicts, which subsequently affects the project delivery process
(Arthur and Pryke 2013). The construction project management system therefore exists to
identify and integrate these micro-differences to facilitate collaborative working practices.
Figure 2.5 is an illustration of systems differentiation within the construction consul-
tant system. The internal parts are linked by a common identity of professionals offering
advice and management services in the construction industry. Within the main system
are multiple subsystems of professional groups including quantity surveyors, sustainabil-
ity consultants, building surveyors, building engineers, architects, fire engineers, building
information modelling (BIM) managers, and project managers. The members within the
quantity surveying subsystem are connected by a common identity of offering professional
services on cost consultancy (RICS 2018a), the members of the sustainability consultant’s
subsystem are linked by a common identity of offering professional advice on sustainable
energy construction (Gov.UK 2021c), the building surveying subsystem is made up of mem-
bers connected by a common purpose of offering professional advice on building pathology
(RICS 2018b), the members within the building engineering subsystem offers professional
advice on construction design and technology (CABE 2021), and the architect subsystem
are made up of professionals who offer architectural design services (RIBA 2021). The fire
engineering subsystem is also composed of members who offer professional advice on the
specification, design, and installation of fire protection solutions in construction projects
20 2 Systems Analysis of the Construction Industry and Project Delivery
Figure 2.4 Conceptualisation of the construction project management system. Sources: Adapted
from Arthur (2018) and Walker (2015).
Interactions
Figure 2.5 Systems differentiation. Sources: Adapted from Carmichael (2006), Cole and Kelly
(2020), and Walker (2015).
2.5 The Construction Project Management System; Differentiation and Risk 21
(IFE 2020). The members within the BIM management subsystem are connected by a com-
mon purpose of integrating building information (Sawhney et al. 2017). Lastly, the project
management subsystem membership comprises of professionals connected by a common
drive to integrate the different specialist roles involved in the construction delivery pro-
cesses, to mitigate conflicts, and cross-purpose working practices (RICS 2018c).
It must be emphasised that despite all the above subsystems being components of the con-
struction consultancy system, and inferably being linked by the common purpose of provid-
ing professional advice and management services, the differences in their micro-objectives
nevertheless, defines the variations in their specific roles and services. The plurality of func-
tions emanating in the network of different professional groups is the result of systems
differentiation.
According to Walker (2007), differentiation within the UK construction industry has been
influenced by the impact of the evolution of the construction disciplines and the nature
of training of the professionals. The other factors which account for differentiation at the
project organisational level includes differences in objectives of parent companies forming
the project team, difference in personal objectives of project team members, and the impact
of prolong environmental changes.
The impact of environmental forces has continued to influence the mutation of the spe-
cialist roles into multiple subgroups. We have also seen the emergence of new specialist
roles to address demands created by new environmental changes. The Royal Institution
of Chartered Surveyors website lists 22 specialist surveying pathways (RICS 2021b). Simi-
larly, a search on the Internet, and academic curriculum of the professional training centres
and universities, confirms the presence of several architectural and engineering disciplines.
There are different specialist architects and engineers for the different construction sec-
tors including residential, restoration, commercial, health, education, and leisure. There
are also specialist architects and engineers for the different construction delivery phases
and stages. Recent promotion of sustainable construction solutions, building information
modelling, and fire safety practices, have also resulted in the appearance of additional spe-
cialist roles including sustainability consultants and energy assessors (Gov.UK 2021c), BIM
managers (RICS 2021c), and fire safety engineers (IFE 2020).
The cumulative impact of the construction environmental changes has been fragmenta-
tion in structure, as seen in the many niche organisations operating within the UK construc-
tion industry (Arthur 2020). The Office for National Statistics recorded 325 736 registered
construction companies in Britain, in 2019 (Allcoat 2018). According to Edkins (2009), the
transition in organisational structure has resulted in a management approach, where many
firms now specialise in aspects of the construction delivery process and rely on input from
the other organisations, to complement their project delivery. And with the variations in
micro-objectives of the different organisations, there has been distinctions in the impact of
the environmental changes and consequently the perception and interpretation of the asso-
ciated risks and opportunities (Japp and Kusche 2008; Arthur and Pryke 2013; Arthur 2018,
2020). The failure to recognise the dissimilarities in the risk communications and designing
appropriate strategies for harmonisation run the risk of introducing conflicting strategies
to impede the construction system’s integration and transformational processes.
2.5.1.3 Project Team Members from Different Organisations and Internal Subgroups
Construction project teams are usually created with professionals drawn from different
organisations of different structures and procedures (Pryke 2006; Winch 2010). Even where
the members come from the same organisation, there may still be differences in culture
between their internal parent subgroupings (Mullins 2005). The assembling together of the
different structures and procedures inherent in the project team members’ parent compa-
nies and subgroups, consequently, creates differentiation.
2.5 The Construction Project Management System; Differentiation and Risk 23
The stages at which project team members are introduced to the project delivery process
could also influence differentiation (Walker 2007). Where the design and construction func-
tions are separated, an early introduction of the constructor to the project fosters integra-
tion with the designers, thereby reducing differentiation. Additionally, during the project
delivery stage, there are instances where additional specialists are introduced to offer an
identified support. If this occurs at a time when the project coalition is well integrated,
then unless the new member is immediately assimilated, there could be potential differen-
tiation. Even where project team members have a previous working experience with each
other, complacency could lead to differentiation, where they fail to realise that the differ-
ences in project environment may demand different transformational approaches (Walker
2007). Chapman (1999) have further suggested that changes in project team membership,
especially the loss of a key member, could lead to differentiation, where the replacement is
not well assimilated into the team.
processes. In such a situation, the project delivery system may be forced to either revise
the original objectives or alter the processes to align. The degree to which a system
can accommodate a sudden environmental change depends on the nature, magnitude,
and frequency of the change and the responsiveness of the internal structure (Walker
2007). Where a system lacks the capacity to adequately respond to an abrupt change, the
ramifications on the internal processes may affect the achievement of the defined project
output (Cagno et al. 2007).
An illustration is the impact of the United Kingdom Coalition Government’s October
2010 spending reviews, which resulted in the sudden suspension and cancellation of public
funded projects. Most public sector clients had to immediately re-structure their business
activities to reduce their expenditure, which consequentially affected their project delivery
programmes, and caused an overall fall in the national construction output (BCIS 2011).
The cumulative effects were escalation in competition among contractors in sourcing out
the few new contracts (Garratt 2010), with most contractors accepting contracts at lower
margins, or even at a loss just to sustain their cash flow (Hammond 2011). The correspond-
ing effects at the project level included protracted client-contractor disputes over contract
details, projects running over budget and programme, and contractors adopting cost reduc-
tion strategies which often compromises the product quality (CBI 2010).
The rapid spread of the coronavirus pandemic has likewise resulted in significant imme-
diate changes to the management and operation of construction activities. The UK govern-
ment’s directives at controlling the rate of infection have included the publication of site
operating procedures for construction activities. This has led to some constructor organ-
isations temporarily suspecting or reducing the capacity of their site activities, with cor-
responding increases in extension of time claims and associated project expenditure. The
promotion of working from home policies have also resulted in virtual project meetings
replacing the typical site meetings. Construction organisations without established infor-
mation technology systems have struggled to respond to the sudden changes, which now
threatens their survival.
the institutional forces which governs the execution of the task. Mullins (2005) adopts the
PESTEL model to expand the environmental forces into multiple differentiated functional
subsystems comprising political, economic, socio-cultural, technological, ecological, and
legal.
The following theoretical discussions review how the bilateral interactions between the
construction system and the multiple social differentiated functional subsystems influence
social construction of risk within the construction system, in the context of risk interpreta-
tion and responses.
of individuals, pressure groups, local and national organisations, and institutions such as
the Greenpeace UK; Foundation for International Environmental Law and Development;
Friends of the earth UK; Extinction rebellion; and the Greta Thunberg activists. All these
groups are consumers of construction products.
It is interesting to note that in recent times, most construction firms are striving to attain
sustainability credentials such as ISO 14001 Environmental Management System and
reduction in carbon footprints. This may partly be due to the increased media awareness
and the need to demonstrate sustainability commitment on most tender selection question-
naires. Government directives also require that public and private buildings become energy
efficient, by being assessed and issued with Displayed Energy Certificates, and Energy
Performance Certificates, respectively (Gov.UK 2021a,b). At the project level, sustainability
has become a mandate for most public funded projects. There is a statutory building
regulation requirement for new build residential developments to achieve minimum 35%
carbon reduction. This has inspired the adoption of energy efficient technologies including
air and ground source heat pumps, photovoltaics (PV) panels, and positive heat ventilation
systems to address the risk of meeting public funding requirements.
2.7 Summary
Chapter 2 has employed systems thinking and analysis in evaluating the construction
industry. The discussions have centred on the structure of the industry, the management
and operational processes, and the interactions with environmental forces, leading to the
social construction of risk.
2.7 Summary 29
The deliberations started with an exposition of the UK construction industry. The Office
of National Statistics publications released on 18 October 2019 confirmed 325 736 registered
construction firms. Construction industry activities excluding the professional services con-
tributed £117 billion and 2.4 million employments to the national economy in 2019. The UK
government’s policy ambitions for the construction industry by 2025 include 33% reduction
in project capital cost and whole life costing, 50% reduction in construction period, 50%
reduction in greenhouse gas emission in the built environment, and 50% reduction in the
trade gap between the exports and importation of construction products and materials. It is
an undeniable fact that the achievement of the policy ambitions will require fundamental
improvements to the construction delivery operational and managerial systems, including
risk management practices.
Taking a systems thinking and analysis approach, the construction industry has been
evaluated as a system receiving human and material resources from its environment and
using industry tools and techniques to convert the materials into construction products
including houses, schools, industries, hospitals, public buildings, and bridges to serve its
environment. Systems thinking is an investigative approach to studying a subject based on
the collective properties of the entirety, rather than the qualities of the individual parts.
The foundational principles and concepts originated from the biological sciences, through
Ludwig Von Bertalanffy’s discovery of general systems theory in the 1960s. Bertalanffy
(1968, 2015) and Arthur (2020) have however argued the pre-existence of systems thinking
application, prior to the formal discovery of general systems theory.
The systems analysis of the construction industry also involved evaluation of key con-
cepts including open and closed systems, systems objectives, systems decomposition, and
systems differentiation. Closed systems operate without the influence of their environmen-
tal forces, whilst opened systems receive inputs from their surroundings to convert into
outputs, which are released back into the environment. The construction industry being an
open system similarly receives inputs in the form of specialist skills acquired from the con-
struction educational institutions and training programmes, plus building and construc-
tion materials produced by the various manufacturers. The inputs are transformed through
the application of industry operational and managerial tools and techniques into houses,
schools, offices, industries, airports, and bridges to serve the construction environment.
The construction system’s objectives are what defines the inputs and transformational
processes required for achieving a desired output. They emanate from the analysis of stake-
holder demands and aspirations, within the framework of clients’ organisational business
cases. Construction system objectives at the strategic level defines broad overall mission
statements and long-term business goals, whereas the operational level objectives express
the required project delivery standards (Carmichael 2006).
Chapter 2 has further analysed the internal structure of the construction industry using
the concepts of systems decomposition and differentiation. The system’s decomposition
analysis has enabled the evaluation of the components within the construction organisa-
tional structure, as lower hierarchy subsystems, comprising of the client organisations, pro-
fessional consultancy firms, and building contractors. The client organisations’ subsystem
has been further analysed into lower order hierarchical structures based on the differences
in market sectors, followed by the differences in firms, departments, programmes, and
projects. The system’s decomposition analysis at the project organisational level centred on
30 2 Systems Analysis of the Construction Industry and Project Delivery
the different transformational processes applied at the project delivery phases, which was
illustrated using the eight stages of the RIBA Plan of Work 2020. Each stage receives inputs
from the preceding stage and the wider construction industry environment to convert into
outputs for the succeeding stages. The conceptualisation of lower order hierarchical struc-
tures exhibiting systems characteristics, implied variations in their micro-objectives, and
consequently, differences in the impact and interpretation of environmental forces.
The theoretical evidence of varied micro-objectives of the different project delivery phases
has also revealed multiple interconnected functionalities, based on the variances in intel-
lectual and emotional orientation of the specialist roles relating to the different phases and
processes, which is termed, systems differentiation. The factors responsible for differentia-
tion within the UK construction industry include the impact of the evolutionary processes
of the different specialist construction disciplines, the application of isolated training pro-
grammes for the various specialist disciplines, the formation of project teams with mem-
bers drawn from different organisations, and internal subgroups, the differences in per-
sonal objectives of project team members, and lastly, the impact of environmental changes
(Walker 2015). The failure to harmonise the variabilities in orientation could potentially
lead to cross-purpose working and conflicts, and if further left unattended, could become
injurious to the achievement of the project main objectives. The project management sys-
tem therefore exists to identify and integrate these differences, to facilitate collaborative
working practices.
Chapter 2 ends with an evaluation of the bilateral interactions between the construction
system and its environmental forces, and how it results in the creation of risk events. The
construction system’s environmental forces have been analysed based on the contempo-
rary structure of differentiated functional political, economic, socio-cultural, technological,
ecological, and legal subsystems. The differences in the differentiated environmental forces
result in differences in impact on the different specialist roles involved in the various project
delivery phases and processes, and subsequently, differences in risk communication.
The next Chapter 3 discusses the concept of risk, including the historical progression of
the risk terminology and the different conceptual interpretations.
31
3.1 Introduction
Chapter 3 discusses risk epistemology using social science theories. It begins with an expo-
sition of the concept of risk from the perspective of the traditional society’s approach to
risk identification and treatment to contemporary thinking of multiple risk communica-
tion. The realisation of inherent risk factors in every human endeavour suggests the need
for careful consideration of the likely benefits and effects of potential future events before
establishing acceptable risk levels. This has been expounded through a review of the risk
management decision-making processes of the ongoing Covid-19 pandemic and the 2017
Grenfell Tower fire incidence in the United Kingdom.
There have been further discussions on risk etymology. The historical review expands
from the playing of games of chance using animal bones in the antiquity world through to
the establishment of insurance facilities in the fourteenth century for shared risk manage-
ment, and finally, contemporary application of the risk terminology in multiples disciplines.
The theoretical review also involves the evaluation of different risk conceptual interpre-
tations. The variances in philosophical basics of the diverse disciplines where risk is applied
have given rise to corresponding variations in conceptual perspectives comprising, real
and objectives, subjectively biased, psychometric, socially mediated, socially constructed,
real and socially constructed, and edgework (Zinn 2008). The psychometric viewpoint as
applied in this book interprets risk in three keyways including risk as a feeling, risk as anal-
ysis, and risk as politics (Slovic et al. 2010). The sociological conceptual perspective also as
expressed in this book proposes risk creation through socio-cultural interactions and impact
from social institutions (Zinn 2008).
agency to change the outcome of a future event is what has theoretically introduced risk
communication into our daily activities. The modern society in effect functions on risk
decision-making.
It has been said that life without risk would be boring. Risk however comes at a price, and
therefore, it is important to evaluate the likely benefits and implications of potential future
events to establish acceptable levels of risks (Zinn 2008). For example, most governments’
approach to the management of the ongoing Covid-19 pandemic has been continuous eval-
uation of the potential benefits and effects of possible interventions to determine appropri-
ate responses. The decision-making processes have involved quantification of the economic,
social, and psychological benefits of relaxing lockdown restrictions against the social cost
of increased infection rates and fatalities. Before arriving at appropriate decisions, gov-
ernments must consider multiple variables including the acceptable rate of infection and
fatality predictions that will justify the relaxation of lockdown and social distancing rules, to
encourage workers to return to their businesses, and to increase economic productivity. For
most western nations including the United Kingdom, the emphasis has been interventions
that contain the rate of infection below R1.
In the context of construction delivery, the 2017 Grenfell Tower fire incident resulted in
the U.K. government evaluating the acceptable level of high-rise building fire risk events
and enacting appropriate changes to the statutory building regulations. The catastrophic
incident presented the government with multiple decision spectrum including the imme-
diate prohibition of combustible building materials in high-rise developments with the
associated immediate expenditure to refurbish the existing housing stock, plus the temporal
stress on accommodation during the decanting process, and the likely increase in devel-
opment cost for new high-rise buildings. The alternatives included accepting the risk and
permitting the status-quo with no changes to the existing housing stock or minimising the
risk through regulation of the use of combustible building materials in defined high-rise
buildings, with associated minimised cost implications. The government after extensive
consultation and risk analysis decided to regulate the remedial process by first stipulating
the use of non-combustible building materials in high-rise buildings over 18 m and subse-
quently, revising to a lower building height.
The risk concept is associated with future events and with the possibility of occurrence.
Events that are certain to occur/not occur fall outside the epistemology of risk. We all go
through life with some level of expectation for the repetition of past events, nevertheless,
until they occur, there will always be a degree of uncertainty associated with our expec-
tations (Zinn 2008). The degree of expectation of repetition of past occurrences can be
presented through a formalised structured approach using statistics and probability or
informally using qualitative statements (Smith et al. 2006; Arthur 2018, 2020).
The application of risk in multiple disciplines has been associated with undesirable nega-
tive events such as hazard, loss, damage, or threat (Zinn 2008). It must however be empha-
sised that risk is not always negative, as the uncertainties associated with future events, can
also result in benefits (Dallas 2006).
Real and objective Risk as a tangible event capable of being Actuarial science,
analysed objectively toxicology, epidemiology,
engineering, probability
analysis, economies
Subjectively biased Risk perceived to be objective but Economies, subjective
subjective judgement could introduce analysis
distortions
Psychometric Risk perception defined by psycho- Psychometric analysis,
physical scaling and multivariate cultural studies
statistics
Socially mediated Risk perception definition, analysis, Cultural studies, sociology
and responses are determined by
socio-cultural institutions
Socially constructed Risk communication constructed by Systems theory,
socio-cultural interactions governmentality
Real and socially Risk as a real event that can also be Risk society, cultural
constructed altered through the influence of studies
socio-cultural institutions
Edgework Appreciates the reality of risk impact Sociology
and associated emotional stimulus
event through the calculation of the probability of occurrence and impact (Zinn 2008).
And where there are difficulties in the calculation, it is interpreted as a requirement for
further cognitive investigations to identify appropriate structured programme to facili-
tate the analysis, rather than an appreciation of the limitations of the risk quantification.
The viewpoint is typically applied in actuarial science, toxicology, epidemiology research,
engineering, probability analysis, and economies disciplines.
Subjectively biased viewpoint: The proponents of this viewpoint not only subscribe to
the principles of objective risk analysis but also believe that the subjective judgements of
risk analysts introduce systematic distortions to their objective calculations (Zinn 2008).
The findings from objective risk analysis are interpreted as ideal responses to pursue,
whilst the risk analyst’s subjective judgements and biases are understood as unreasonable
responses. The perspective is typically applied in economies, where risk is interpreted as
utility to be maximised through rational judgement and that the influence of subjective
preference misrepresents the ideal outcome.
influence of affect emotions, mood, heuristics, on perception formation, and risk judgement
(Slovic 2010; Kahneman 2011). There have also been other research studies analysing the
impact of culture and emotion on risk communication (Tulloch 2008; Slovic 2010; Bateman
et al. 2010).
and interest amid the constant changes in the social setup is what creates the risk communi-
cation (Japp and Kusche 2008). The systems thinking approach makes it possible to engage
in abstract social risk dialogue, without reference to the real world. This has been praised for
enabling governments to pre-evaluate the likely implications of intended policies to ascer-
tain acceptable impact levels before their implementation. Notwithstanding, it has equally
been criticised for enabling rhetoric analysis that sometimes bear no relationship to reality.
which is the mediation of the conflicts observed in the clashing of the traditional subjective
methodology, and the modern scientific approach (Slovic et al. 2010). The sociological risk
analysis, on the other hand, has employed systems thinking and analysing in modelling the
construction industry, its internal structures, and the processes of the project delivery, and
management systems.
Loosemore et al. (2006) in taking a sociological conceptual approach has described risk as
an uncertain event that might occur in the future to potentially affect an interest or objective
(usually adversely), although the precise likelihood and impact may be indeterminate. The
key theoretical implications are the relationship between risk and uncertainty, risk as a
future event, and the potential for risk to affect the interests/objectives of the related subject.
Loosemore et al. (2006) have argued that the absence of information makes it difficult to
analyse potential future events, thereby creating uncertainties about the likely impact and
required remedial actions. Edkins (2009), on the other hand, has argued theoretical differ-
entiation between ‘risk’ and ‘uncertainty’. Taking the real and objective viewpoint, Edkins
has described risk as a potential future event which is tangible and able to be analysed,
whilst uncertainty exists in the peripheral and not concrete enough to warrant objective
analysis. A careful evaluation of the different expositions offered by Loosemore et al. (2006)
and Edkins (2009) confirms the influence of the epistemology of their different approaches
to risk conceptualisation.
The second theoretical implication is the description of risk as a future event. Loosemore
et al. (2006) have postulated that risk analysis must be based on the events that give rise to
risk outcomes, rather than the potential impacts or consequences. Accepting this exposition
in effect makes it wrong to categorise risk based on the probable implications on project
objectives, which is seen as reactive, and therefore, more in line with ‘crisis management’
rather than ‘risk management’.
Risk is also associated with potential future events, rather than past and present
events. Aven (2010) has however argued against the conceptualisation of risk as an event
on account of such approach limiting the ability to analyse risk. He has subsequently
suggested risk evaluation and quantification based on the antecedent uncertainties which
result in the occurrence of risk events.
The last theoretical implication is the need to establish the potential for a future occur-
rence to affect the interests or objectives of a subject before classifying it as an associated
risk event. This suggests that not all future events can be classified as risk to all subjects.
There must first be an establishment of potential impact (Loosemore et al. 2006). It is pos-
sible for a future event to be perceived as risk by the organisations and individuals likely to
be affected by its occurrence and at the same time become unnoticeable to the remaining
population. The differences in personal and corporate interests and objectives also suggest
differences in impact of potential future events, and subsequently, differences in appre-
ciation of potential risk events (Arthur and Pryke 2013; Arthur 2018, 2020). When the
Covid-19 issue started in the late 2019, the public discourse then was limited to the few
Asian countries that were then experiencing the impact. And as the infestations and fatali-
ties degenerated into a global pandemic, with the extension of socio-economic impact into
other countries and industries, there has been corresponding extension of the public risk
discussions with multiple nations and industries developing comprehensive Covid-19 risk
management plans.
38 3 The Concept of Risk
3.6 Summary
Chapter 3 has taken a theoretical review of the risk concept using social science theories.
Risk is seen as being part of human existence. The events that threaten human survival were
traditionally identified and addressed through the guidance and protection of hierarchical
social religious systems. These were in the form of guided morality, superstitions, taboos,
and rituals. Contemporary social structures, however, shows the displacement of the hier-
archical religious systems by several functional specialist systems operating on identical
horizontal levels. There is now the belief in human ability to alter the negative effects of
social events, rather than simply waiting for supernatural interventions. This has subse-
quently extended risk communication into the public realm (Japp and Kusche 2008).
Risk management decision-making involves prior evaluation of the likely benefits and
effects of potential future events to determine the acceptable risk levels. This has been evi-
dent in the management of the ongoing Covid-19 pandemic, where the decision processing
of many governments has involved prior evaluation of the economic, social, and psycholog-
ical benefits of easing restrictions, against the likely impact on infection rates and fatalities.
Similarly, the UK government’s policy and regulatory changes after the 2017 Grenfell Tower
fire incident followed public consultations and examination of the likely benefits and set-
backs of different remedial interventions.
The origin of the concept of risk has been associated with the emergence of games of
chance in the antiquity world, using a six-sided animal bone called ‘astragalus’. The risk
conceptualisation during the early medieval period centred around the awareness of com-
munal threats, which were identified and managed through the guidance and protection of
the hierarchical religious establishments. The risk terminology then was related to the fear
of suffering physical or psychological injuries from pirate attacks or hazardous conditions
during voyages (Romeike and Hager 2009). The fourteenth century saw the introduction
of insurance policies in the northern Italian city states for the shared management of the
risk associated with shipwreck. There was also the application of probability techniques for
risk quantification (Risk Net 2020). The risk terminology subsequently became a common
vocabulary in other European commercial industries (Zinn 2008), appearing in England in
the 1830s when it was used in the insurance industry (Smith et al. 2006).
The differences in epistemology of the varied disciplines where risk terminology is
applied is what has given rise to the differences in conceptual interpretations. The real
and objective perspective as practised in the actuarial sciences, toxicology, epidemiology,
engineering, probability analysis, and economies disciplines, understand risk to be a
tangible event capable of being analysed objectively. The subjectively biased perspective as
applied in economies and subjective analysis disciplines, perceives risk to be objective and
at the same time, susceptible to distortions from the influence of subjective judgements.
The psychometric perspective utilised in psychometric analysis, cultural studies, and this
book defines risk perceptions through psychophysical scaling and multivariate statistics.
The socially mediated perspective as practised in sociology and cultural studies believe risk
perceptions, analysis, and responses are defined by socio-cultural institutions. The social
construction perspective as applied in systems theory, governmentality studies, and in this
book understands risk communication to be a construct of socio-cultural interactions. The
real and socially constructed perspective utilised in Ulrich Beck’s analysis of risk society
3.6 Summary 39
and in cultural studies believes risk to be a real event that can be altered through the
influence of socio-cultural institutions. Lastly, the edgework perspective, as explained by
Stephen Lyng, appreciates the reality of risk impact and the emotional stimulus associated
with high-risk activities (Lyng 2008).
The psychometric risk perspective application in this book has centred on the interpreta-
tion of risk as a feeling generated through heuristics-inspired personal perceptions to guide
impulsive responses to threats; risk as analysis taking into consideration the sensitivity of
subjective variances; and risk as politics mediating the systematic conflicts between the sub-
jective and rational approaches. The sociological risk perspective also utilised in this book
has applied systems thinking and analysis in modelling the construction industry, the inter-
nal structure, project delivery, and project management processes. Theoretical evaluation
of the risk concept as explained by Loosemore et al. (2006) has revealed conceptual rela-
tionship between risk and uncertainty; risk as a future event; and the need to first establish
potential impact of a future event on the interest/objectives of a subject before classifying
as a risk event.
The next Chapter 4 extends the sociological risk conceptualisation by using systems
thinking and analysis to model the construction risk management process and sub-stages.
There is also theoretical review of the construction risk management approaches.
41
4.1 Introduction
Chapter 4 discusses the construction risk management process. The review adopts a socially
constructed risk perspective in modelling the construction risk management stages and
processes using systems thinking and analysis. There is also theoretical review of the dif-
ferent risk management approaches comprising the traditional approach of intuition and
contemporary application of systematic and scientific techniques.
The chapter begins by reviewing the changing perspectives of organisational risk man-
agement strategies. The traditional approach of transferring risk management responsibili-
ties to third parties using insurance and contracts seems to be fading away. The impact from
the latest global economic recessions and associated market instabilities coupled with natu-
ral catastrophic events, terrorist incidence, global pandemics have collectively exacerbated
the social risk barometer. The volatility of socio-economic order has subsequently increased
insurance premium thereby making the reliance on insurance as a corporate strategy for
managing risk non-viable.
The modelling of the construction risk management process has involved systems inter-
pretation of the key interconnecting stages comprising of risk identification, risk analysis,
risk response, and risk review. The concept of systems decomposition has subsequently
been applied in analysing the different risk management stages as subsystems displaying
general systems properties of input, transformation, and output. The subsequent evalua-
tion of the bilateral interactions between the internal processes at the different stages and
the wider construction project and industry environment results in the modelling of the
construction risk management system.
There has been further analytical review on the different risk management approaches:
intuition versus rational techniques. The theoretical discussion centres on the benefits
and downsides of the different methodologies and subsequently, the analytical argument
for incorporating both approaches to facilitate effective risk management systems. The
behavioural sciences explanations for the different decision settings inspiring the selection
of an appropriate risk management approach have also been discussed. The chapter
ends with a review of the construction management publications, which reveals high
incidence of intuitive construction risk management practices, by professionals with
limited knowledge and understanding in the rational decision-making techniques.
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
42 4 Construction Risk Management
There are multiple models that have been developed for describing the risk management
process. The association of project managers have proposed a five-stage process involving
initiation, identification, assessment, planning responses, and implementation. Smith et al.
(2006) have also suggested a four-part process comprising of risk identification, risk anal-
ysis, risk response, and risk review. Dallas (2006) believes the different stages operate as a
cycle of interconnected parts, as illustrated in Figure 4.1.
Using the principles of systems thinking and analysis which were discussed in Chapter
2, we can analyse the different stages of the risk management process as sub-systems pos-
sessing input, transformation, and output phases. The concept of systems decomposition
suggests the different internal phases will function as sub subsystems of the different risk
management stages, and will require interaction of the parts to facilitate achievement of
their micro-objectives (see Figure 4.2).
The previous discussions on the construction industry being an open system implies
the operation of the risk management stages will also involve interactions with the other
systems and subsystems within the construction project management framework and the
wider construction industry environment. This has been depicted in Figure 4.3.
The multiplicity of tools, techniques, and approaches applied within the different stages
and phases coupled with the intricacies of interactions within project internal structures
and the variability of the construction industry environmental factors confirms the con-
struction risk management process to be complex and dynamic. There is therefore the need
Risk identification Risk analysis
Risk review Risk response
Figure 4.1 Risk management cycle. Sources: Adapted from Smith et al. (2006) and Dallas (2006).
44 4 Construction Risk Management
Risk identification subsystem Risk analysis subsystem
Risk review subsystem Risk response subsystem
Figure 4.2 Risk management subsystem. Sources: Adapted from Walker (2007, 2015), Smith et al.
(2006), Dallas (2006), and Arthur (2020).
for constant monitoring and adaptations in response to the continuous internal and envi-
ronmental changes (Arthur 2020).
Construction risk
management
environment
Risk identification subsystem Risk analysis subsystem
I T O I T O
Risk review subsystem Risk response subsystem
I T O I T O
I: Input
T: Transformation
O: Output
: Sequences
: Interactions
Figure 4.3 Construction risk management system. Sources: Adapted from Walker (2007, 2015),
Smith et al. (2006), Dallas (2006), and Arthur (2020).
the establishment of strengths and opportunities to enhance and maintain, and weakness
and threats to control or avoid in the future application of the model. The robustness of
the risk review stage is enhanced through the wide involvement of the specialist roles and
organisations contributing to the risk management process.
1. Holistic 1. Analytic
2. Affective: pleasure-pain oriented 2. Logical: reason oriented (what is sensible)
3. Associationist connections 3. Logical connections
4. Behaviour mediated by ‘vibes’ from past 4. Behaviour mediated by conscious appraisal
experiences of events
5. Encodes reality in concrete images, 5. Encodes reality in abstract symbols, words,
metaphors, and narratives and numbers
6. More rapid processing: oriented towards 6. Slower processing: oriented towards delayed
immediate action action
7. Self-evidently valid: ‘experiencing is believing’ 7. Requires justification via logic and evidence
et al. (2010) have suggested that while events which evoke strong affect reactions such as
terrorism, infatuation, faith, requires cognitive analysis to inject some reality into the risk
judgement and decision-making, probability predictions of events involving fatalities may
fall short of conveying the potential danger and hence requires affect descriptions to ensure
rational risk judgement.
According to Inbar et al. (2010), the characteristics of a decision task and its envi-
ronment are the key factors that influences the choice of a decision-making approach.
Thus, where the decision task and environment exhibit characteristics associated with
reasoning, rationality will be applied, and likewise, in a situation where the key exhibits
of the decision task and environment relates to ‘feeling’, then intuition will be applied.
Traditionally, the rational approach has been widely agreed by scholars as the appropriate
model for decision-making (Inbar et al. 2010). Other research findings, however, suggest
that not everyone follows the norm of rationality when it comes to judgement and
decision-making (Finucane et al. 2003). Zajonc (1980) and Slovic et al. (1981) have argued
that decision-making in real-life situations is guided by both perception and rationality.
Even when we think we are making decisions purely based on rationality, we are in effect
using rational adjustments to refine an initial anchored position which is defined by
background perceptions (Tversky and Kahneman 1982a; Kahneman 2011).
Research findings in construction risk management reveals high reliance on perceptions.
A questionnaire survey of 100 contractors and consultants in the UK construction indus-
try, revealed the risk management practices being dominated by people without formal
training in risk management systems who relies on intuition and heuristics for their anal-
ysis and decision making (Akintoye and Macleod 1997). Lyons and Skitmore (2004) after
analysing 44 received responses out of a sample of 200 engineering companies in Queens-
land, Australia, identified similar findings of high incidence of intuitive risk management
practices, carried out by people with minimal formal training in risk management tech-
niques. Another questionnaire survey of 84 sampled construction contractors in Malawi
also revealed that apart from the small proportion of large and experienced companies
48 4 Construction Risk Management
Table 4.2 Reasons for low application of rational construction risk management techniques.
Year of research
publication Reasons for low application of rational risk management techniques
Kululanga and ● The findings of the study identified low application of rational techniques to
Kuotcha (2010) be most typical of the small-sized and less-experienced construction firms
(There was no specific survey question to solicit the reasons for the low
application of rational techniques)
Sources: Adapted from Akintoye and Macleod (1997), Lyons and Skitmore (2004), and Kululanga and
Kuotcha (2010).
majority of the small and medium companies did not apply formal risk management tech-
niques (Kululanga and Kuotcha 2010).
Among the reasons that have been given for the low application of rational risk manage-
ment practices in the construction industry are the unavailability of data to enable ratio-
nal analysis, lack of competency in rational risk management systems, size and scope of
most construction projects do not justify the cost and time required for applying rational
techniques, and non-appreciation of the need for scientific tools in construction risk man-
agement (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha
2010 – see Table 4.2). A major limitation on the validity and conclusiveness of the above lists
is that they are merely a compilation of survey responses and do not include an adequate
evaluation of the psychological factors which leads to intuitive decision-making.
From the psychological point of view, Slovic and Peters (2006), Bateman et al. (2010),
and Kahneman (2011) have suggested that perceptions tend to occur spontaneously and
4.5 Summary 49
is rapidly available making it easier and faster to employ in decision-making than relying
on rational analysis, especially where complex decisions are required, and scientific data is
limited. Other studies also suggest that in practice, people do not trust, understand, or use
probability estimates in their risk management decisions (March and Shapira 1987; Akin-
toye and Macleod 1997; Edwards and Bowen 1998; Lyons and Skitmore 2004). It should
however be noted that the psychological findings have been research from mainly the gam-
bling industry, which although similar in some ways to the construction industry in that
each project is bespoke yet exhibits differences in the nature and time scale of required
decision-making.
4.5 Summary
Chapter 4 has reviewed the construction risk management process. Theoretical evidence
confirms the preference of most managers to transfer their risk events to third-party organ-
isations, using insurance and contract provisions, rather than managing themselves. The
shortcoming of this approach is that sometimes risk is passed on to lower-tier organisa-
tions, without the capacity to manage, and thereby introducing additional risk to the project
delivery. The recent global events including economic recessions with high market volatil-
ity, and liquidation of established international organisations, terrorist incidence, natural
upheavals including hurricanes, earthquakes, wild bush fires outbreaks, domestic fires and
fatalities in high rise buildings, greenhouse gas emission and depletion of the ozone layer,
have cumulatively increased insurance premiums, thereby making the reliance on insur-
ance policies as a risk management strategy non-viable. In addition, government’s reg-
ulatory activities in checking corporate risk management systems and the requirement
to demonstrate risk management competencies in public procurement have resulted in
a changing perspective on corporate risk management, with more emphasis now being
placed on internal management strategies.
Taking a socially constructed risk conceptualisation, the construction risk management
process has been modelled using systems language. The different interconnecting stages
comprising risk identification, risk analysis, risk response, and risk review have been anal-
ysed as sub-systems using the concept of systems decomposition. The risk identification
sub-system being the first stage of the process receives input from the internal project
structure and the wider construction industry environment to determine the potential
events that could affect the project success. The risk analysis sub-system then evaluates
reliability of the identified risk events and further categorises based on their likelihood of
occurrence and impact. The risk response sub-system evaluates the appropriate manage-
ment response. The risk review sub-system completes the risk management process. It is
responsible for auditing the functionality and effectiveness of the applied methodology.
The internal structure of the construction risk management process subsystems has
revealed micro-objectives guiding their input, transformation, and output phases. The
construction industry being an open system, also suggests bilateral interactions between
the internal structure and the wider environment. The assortment of tools and techniques
utilised within the different phases and stages, in addition to the fluidity of environmental
50 4 Construction Risk Management
forces makes the construction risk management model very dynamic and responsive to
the continuous project and industry changes.
The chapter concluded with a theoretical review on risk management approaches com-
prising of intuition and rationality. The intuitive vehicle relies on personal perceptions gen-
erated through heuristics. The rational methodology however involves scientific analysis
of structured data. Other recent studies have discovered the use of fuzzy logic in process-
ing risk data expressed in the form of linguistics variables. Fuzzy logic has been argued as
another form of the rational risk management approach with the differences being the data
presentation format (Arthur and Pryke 2014). The selection of a risk management approach
is influenced by the decision-making setting. Decision tasks associated with feeling thrives
under the intuitive approach, whilst decision tasks relating to reasoning are effectively anal-
ysed using the rational approach. The inherent limitations of the approaches suggest the
benefit of combining principles from intuition and rationality to facilitate effective risk
management system. The prevailing construction management research publications have
however revealed dominant intuitive risk management practices by professionals with lim-
ited competency in rational decision-making techniques. A questionnaire survey of 100
contractors and professional firms in the UK construction industry discovered high inci-
dence of instinctive risk management practices, by professionals without formal training in
risk management systems (Akintoye and Macleod 1997). Another research of 200 engineer-
ing companies in Queensland, Australia, also returned similar finding of dominant intuitive
risk management practices carried out by professional with minimal training in scientific
decision processing techniques (Lyons and Skitmore 2004). Further questionnaire survey
of 84 sampled construction contractors in Malawi also revealed limited application of struc-
tured risk management techniques by the few large and experienced companies, with the
remaining small and medium companies resorting to informal practices (Kululanga and
Kuotcha 2010).
The empirical reasons expounded for the low application of rational risk management
techniques include the unavailability of structured data to compliment the scientific anal-
ysis, low competency in rational decision processing programmes, the value and duration
of construction projects being too low to justify the required financial investment in struc-
tured risk management systems. A major limitation to the accuracy and decisiveness of the
empirical reasons has been the fact that, they appear to be a record of the research findings
without adequate analytical evaluation. The next Chapter 5 will therefore apply the psycho-
metric risk conception to analysis the behavioural reasons underlying the current intuitive
construction risk management practices, and the wider implications.
51
5.1 Introduction
Chapter 5 employs a psychometric risk perspective to evaluate the prevailing intuitive risk
management practices and the wider theoretical and practical issues. The discussion begins
with a review of the different systems of thinking and decision-making. The quick approach
applied in intuitive risk management is always active and constantly generating percep-
tions to guide decision-making. The gradual approach associated with rational risk man-
agement, on the other hand, requires deliberate activation and application of appropriate
cognitive programme to systematically process a decision setting. Behavioural science the-
ories confirm constant interactions between the two systems. Decision processing usually
commences within the quick approach using heuristics to evaluate the prime attributes of
the decision task, for evidence of similar previous experiences to influence the modelling of
the projected outcome in similar patterns. Where there are no comparable previous records
the decision task is transferred to the gradual approach to process using a suitable cognitive
programme (Kahneman 2011).
The chapter then proceeds to undertake deeper theoretical review of the quick
decision-making approach associated with the current intuitive risk management prac-
tices. This has involved analytical review of the psychology of personal perceptions
responsible for intuitive decision processing. The theoretical review has centred on the
key determinants and processes for generating personal perceptions and intuitive decision
processing. The impact of personal perceptions on cognitive reasoning has also been
evaluated. There has been further review of the different risk data presentation formats of
quantitative and qualitative, and how the historical development of the risk terminology
has influenced risk data presentation in statistical and probability formats even within
the construction industry with the dominant intuitive practices. The theoretical issues
involved in the intuitive processing of statistics and probability risk data formats within
the construction industry have been examined.
There has been further discussion on risk management decision-making under intu-
ition. The findings from behavioural science research confirm the susceptibility of intuitive
decision-making to biases and systematic errors. Intuitive decision processing using the
representative heuristics become prone to biases resulting from misconception of the law
of chance, insensitivity to prior probability, sample size, reliability, and validity of base
information. The errors of intuitive decision processing under the availability heuristics
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
52 5 Construction Risk Management Decision-Making
results from how easily past events and searches comparable to current decision tasks can
be retrieved. Intuitive decision-making under the ‘anchor and adjustment’ heuristics also
produces errors, when an assessor misinterprets an anchored value to be close to the final
estimate, leading to pre-matured abrogation of the adjustment process.
The chapter concludes with an evaluation of the sources of risk perception categorisation.
This is on the basis that a correct understanding of the sources of the different personal
perceptions influencing the intuitive construction risk management practices will assist the
project management process to identify and harmonise towards the achievement of project
objectives, and mitigation of cross purpose working practices. There has therefore been a
review of the present theoretical explanations for risk perception categorisation including
personality traits, prospect theory, differences between external stakeholders and project
team members, and cultural theory. The criticisms and limitations of the existing theories
lead to the proposition of a novel theory based on the processes of systems decomposition
and differentiated discussed in Chapter 2, and the conceptual relationship between a ‘risk
event’ and its ‘impact on an objective’ previously discussed in Chapter 3.
Behavioural scientists believe there are two systems of thinking and decision-making.
Various psychologist over the years have assigned different labels for the classification
including, Kahneman (2011) description of ‘fast system’ and ‘slow system’ (Kahneman
2011), and Stanovich et al. (2016) use of ‘system 1’ and ‘system 2’. This book on the other
hand, has adopted ‘Quick approach’ and ‘Gradual approach’ to delineate the two systems
(see Figure 5.1). The primary differences between the systems are the quick approach is
always active and operates spontaneously without effort. The gradual approach however
requires a conscious deliberate effort to activate and sustained attention for the decision
processing.
Decision processing usually commences within the quick approach through involuntary
activation by comparing the decision task against records of previous experiences to
see if there are similarities to guide in the modelling of the decision outcome to reflect
patterns of previous occurrence (Benthin et al. 1993; Damasio 2006). Where there are no
similarities the gradual approach is recalled through voluntary activation of an appropriate
cognitive programme to process the decision task. Thus, the gradual approach becomes
fully activated whenever a person encounters a decision task which exhibit properties
outside the operational world of the quick approach.
Figure 5.1 The two systems of thinking and decision-making. Source: Adapted from Kahneman
(2011).
5.2 The Two Systems of Thinking and Decision-Making 53
Decision-making
setting:
a. Data: Availability of
appropriate data –
record of similar
previous experiences.
b. Approach:
Application of
appropriate qualitative Quality of decision
Spontaneity
approach to analyse outcome
patterns between the
decision task and the
record of previous
experiences.
c. Competency:
Decision maker being
physiologically capable
of instinctive
perception generation.
Figure 5.2 Quick decision-making approach. Source: Adapted from Kahneman (2011).
54 5 Construction Risk Management Decision-Making
The spontaneity of the quick approach enables swift and accurate simple decision-making
in the face of an emergency. Notwithstanding, the difficulty in taming the thought gen-
eration process sometimes results in impulsive and nuisance judgements and project
actions. A major criticism of the quick approach has been the inherent biases which makes
it susceptibility to systematic errors (Tversky and Kahneman 1982a,b,c; Kahneman and
Tversky 1982b; Kahneman 2011). It often associates correct responses with events that
bear resemblance with past occurrence and those easy to construct into patterns. It has
also been criticised for lacking understanding in logic and statistics.
Decision-making
setting:
a. Data: Availability of
appropriate
quantitative data.
b. Approach:
Application of
appropriate rational
approach to Quality of decision
Attention
decompose the outcome
decision task into
structured logical
sequence.
c. Competency: Ability
of the decision maker
to understand and
apply the rational tool
in analysing
quantitative data.
Figure 5.3 Gradual decision-making approach. Source: Adapted from Kahneman (2011).
5.3 The Psychology of Perception 55
The above activities require deliberate activation of suitable rational programmes and
sustained attention for the decision processing. Kahneman (2011) and Bikart (2019) have
suggested that manoeuvring through daily life activities involve allocation of attention to
different rational tasks. In a situation where an activity requires more attention than what
has been deposited, the deficit often results in disruption, and subsequent minimisation of
the quality of the decision output or outright abrogation of the process. This explains why
it is often difficult to undertake multiple reasoning activities at the same time. Kahneman
(2011), believe we all have some level of awareness of the limited capacity of attention. As an
illustration, most adults living with other professionals who work from home often would
become quiet and less disruptive whenever there is a work-related telephone call to allow
the professional the necessary tranquillity to concentrate on the deliberations. Chabris and
Simons (2010) have explained the theoretical basis to be the temporary blindness to sec-
ondary interactions whenever a person become intensely absorbed in a task.
The theoretical evidence of the gradual approach requiring attention for decision process-
ing confirms a direct relationship between the level of provided attention and the quality
of the decision outcome. Being alert during a gradual decision processing will invariably
result in corresponding value to the decision output. The other evidence on the difficulties
in achieving maximum concurrent attention for multiple tasks also suggest that reduc-
ing decision-making to few variables will result in corresponding increase in efficiency.
According to Bikart (2019), Facebook CEO Mark Zuckerberg once explained the reason for
adopting his trademark fashion of grey T-shirt and hoodie was to remove fashion variables
from his daily decision-making, and focus his attention on his core business tasks.
Research findings suggest there are constant interactions between the quick and grad-
ual approaches. The quick approach with its hyperactive properties routinely scans the
constructs within the environment to produce suggestions in the form of insights, convic-
tions, intuitions, feelings, opinions, and impressions for the gradual approach to analyse
and where validated convert them into credence and voluntary actions (Kahneman 2011).
Other research findings also suggest that prolonged practice of cognitive activities asso-
ciated with the gradual approach often become assimilated into practices under the quick
approach (Kahneman 2011). The theoretical explanation is; repeated practice of a mental
activity creates perceptual images and stimulus within the human mental library to influ-
ence intuitive decision processing (Damasio 2006; Benthin et al. 1993).
Research findings within the Affect theoretical framework have identified a direct rela-
tionship between affect and behaviour patterns including perceptions (Zajonc 1980; Forgas
2001; Dohle et al. 2010). Affect as explained by Bateman et al. (2010) expresses an experi-
enced feeling state triggered by a positive or negative inducement.
Affect heuristics, on the other hand, is the experiential process through which positive
and negative affective feelings generate perceptions to guide decision-making (Finucane
56 5 Construction Risk Management Decision-Making
et al. 2003). Slovic et al. (2010) have also suggested a relationship between affect heuristics
and the representative and availability heuristics. The representative heuristic, as explained
by Kahneman and Tversky (1982a), Tversky and Kahneman (2002), and Kahneman (2011),
judges the likelihood of association between an event and its parent population by the
degree to which the event exhibits the prime characteristics of the parent population.
Kahneman and Tversky (1982b) have also suggested that representation exist in situations
where the composition of a sample reflect the majority–minority composition of its parent
population and where the sample composition reflects a possible outcome from the
composition of the parent population. Availability heuristic, on the other hand, predicts
the frequency of an event by how easily a similar past occurrence can be recollected or a
possible outcome scenario can be constructed from the event (Kahneman 2011).
There are different theoretical interpretations for the relationship between affect and the
other physiological feeling states such as ‘emotion’ and ‘mood’. Finucane et al. (2003) refer
to ‘emotion’ as specific intense feeling based on an antecedent cause and normally short
lived. ‘Mood’ on the other hand is explained as a salient feeling state, which may not have
an antecedent cause and may last from a short period to a long period. Isen (1993) cited in
Finucane et al. (2003) refer to affect as the ‘background mood’. Finucane et al. (2003) also
postulates that affect lies between ‘emotion’ and ‘mood’ having an antecedent cause and
normally expressed in a non-elaborate form. Forgas (1995), on the other hand, sees affect
to comprise of both ‘emotion’ and ‘mood’.
Larsen and Diener (1987) and Patrick (1994) have contended that affect experiences
depend on an individual’s psychological composition which influences the ability to rec-
ollect and process past events. Peters and Slovic (2000) argues that optimistic individuals
are better at processing information likely to lead to positive outcomes, while anxious
individuals tend to be better at processing information likely to assist in avoiding a negative
outcome.
with the subject. This can be illustrated by examining Sir Ian Blair’s report in 2008 that the
UK police between 2000 and 2008 identified and foiled 15 terrorist plots in Britain (Slack
2008). The Heritage Foundation also reported that from 11 September 2001 to 11 September
2011 there were 40 terror plots against the United States of America that were identified and
foiled (Carafano and Zuckerman 2011). The identification and foiling of all these terror plots
were barely noticed by the regular public due to their limited knowledge and exposure to
techniques in identifying terrorist plots.
C
A B
Positive image No previous image
Negative image
Figure 5.4 How perceptions are formed. Sources: Adapted from Bateman et al. (2010), Benthin
et al. (1993), and Finucane et al. (2003).
Outcome ‘B’ is where the previous experience with timber frame construction had
occurred on a project where there was a tragic incident with catastrophic consequences,
like fire destructions. The stored negative image induces a negative feeling which leads
to the formation of a negative image on the future event. This ultimately leads to the
development of a negative perception and assessment of timber frame construction
solution as a negative risk to the future project.
Outcome C on the other hand is a situation where there is no previous academic or
working experience with timber frame construction. The assessor does not have any stored
mental image of a previous timber frame project and therefore unable to form risk percep-
tion on timber frame construction.
According to Zajonc (2001), repeated exposure to an event evokes positive affect about the
event. This suggests that where an individual’s previous exposures to an event had occurred
in a neutral context, the mere repeat exposure to the same event could influence the mod-
elling of a positive perception on a similar future event.
Finucane et al. (2003) have also indicated that in a situation where the mental image
library produces multiple positive and negative images as a result of multiple previous
exposures to similar experiences then factors such as the dominant properties of the previ-
ous exposures, and the physiological state of the decision-maker being either ‘extrovert’ or
‘introvert’ will cumulatively influence the modelling of the future event as either a positive
or negative image.
5.3 The Psychology of Perception 59
According to Fischhoff et al. (1978) and Alhakami and Slovic (1994), there is an inverse
relationship between perceived risk and perceived benefit. Finucane et al. (2003) further
suggest that reduction in the time allowed for decision-making increases the inverse rela-
tionship. This can be seen to be contrary to real life where high-risk events are often asso-
ciated with high benefit results (Slovic and Peters 2006).
In addition, Alhakami and Slovic (1994) note that exposure to additional contrary infor-
mation of an event at the time of a risk judgment may alter previous perceptions even
where the base situation has not changed. This may be due to the impact of perceiving the
additional information as a case of suggestion (Kahneman 2011, p. 122) and generating asso-
ciated evidence to confirm the essential characteristics of the additional information. The
consequences often are wrong decisions which affect project success. Flyvbjerg et al. (2003)
in examining the causes of failure of mega projects in Europe identified that on projects
including the Channel Tunnel between France and United Kingdom, the Great Belt Link at
Denmark, and the Oresund Link between Sweden and Denmark, unrealistic best-case sce-
nario predictions presented at the feasibility stages influenced the perception of the project
sponsors into judging the project to be financially viable and approving their implementa-
tion only for the actual project cost to exceed the initial estimates. Likewise, the National
Audit Office (2000) official review of the Millennium Dome project in the United Kingdom
also identified that unrealistic fee-paying visitor projections used at the initial project stages
to justify the project financial viability was not achieved after completion and consequently,
contributed to the project failure.
2015) suggest that mixing approaches and data formats from different decision-making sys-
tems may introduce conflicting variables which may be detrimental to the risk management
process.
According to Tversky and Kahneman (1982a), the factors which cause biases in judge-
ment under the representative heuristics include prior probability, sample size, misconcep-
tion of the law of chance, predictability, and validity. Tversky and Kahneman (1982a) have
further suggested that judgement under the availability heuristics may produce errors as a
result of biases resulting from the ease of retrieving past instances, and the ability to search
past occurrences.
Prior probability is the predicted probability of an event based on its evaluable subjects
within a given population. Thus, given the statistical data of the causes of cost overrun
observed in 100 projects as 40 incidence of design changes; 32 incidence of wrong initial
estimates; and 28 incidence of fiscal changes, the ‘prior probability’ for cost overrun result-
ing from design changes in a similar project become 40%. Kahneman and Tversky (1973)
discovered in an occupation classification experiment that when personality information
on the subjects were provided the assessors became insensitive to the prior probabilities
and rather concentrated on making their predictions by evaluating the similarities between
the provided personality information, and their pre-convinced stereotypical characterises
of the various occupation, which may not be correct.
Tversky and Kahneman (1982a) have also suggested that probability predictions under
the representative heuristics may become insensitive to sample size leading to false expec-
tation of similarities in composition and arrangement between the sample population and
the parent population. This is inconsistent with the tenets of sample theory which does not
guarantee fairness in drawing samples out of their parent population. Tversky and Kah-
neman (1982b) have suggested that unlike the law of equilibrium which corrects errors of
deviation with corresponding counter deviations, the law of chance corrects errors of devi-
ation through dilution. This means that the larger the sample size, the more likely it will
reflect the prime characteristics of its parent population (Kahneman 2011). Intuitive risk
management practices which identifies and treats potential risk events based on few previ-
ous similar observations therefore run the risk of producing errors.
Another potential error that may result from risk management analysis utilising the rep-
resentative heuristics is the misconception of chance. Tversky and Kahneman (1982a) have
explained that most people who make risk management decisions based on the represen-
tative heuristics expect balanced systematic repetition of past occurrence, but this may be
contrary to real-life experiences where chance occurs in no systematic order.
The error of predictability occurs where assessors become insensitive to the reliability of
the base information upon which predictions are made (Tversky and Kahneman 1982a). In
situations where predictions are made based on information which do not relate directly
to an event, then the potential future outcome may exhibit no resemblance to the base
information. Within the construction industry, intuitive risk management decisions which
attempts to predict project success using base information generated under different success
indicators may result in errors.
According to Slovic et al. (1982), the source and presentation format of data can alter
understanding and validity. Therefore, in a situation where an assessor makes intuitive risk
5.4 Risk Management Decision Making Under Intuition 63
management decisions based on propaganda data, rather than facts, there is bound to be
errors.
The principles of the availability heuristics restrict intuitive prediction of potential future
events to past occurrences that can be easily retrieved. Future predictions therefore become
possible only where related past occurrence can be easily recalled. According to Tversky
and Kahneman (1982c), past events associated with famous subjects and celebrities are
usually assigned high probability predictions of association with similar potential future
events compared to the regular occurrences. The inherent limitations in effect make the
availability heuristics ineffective in addressing emergent and peripheral risk events.
Furthermore, the degree of involvement in a past event may influence the ability to eas-
ily recall the outcome. A person’s recollection of an event where there had been a direct
involvement will likely be higher than where there had been an indirect involvement, such
as merely reading about the event. Slovic et al. (1982) have also suggested that events that
stay in the media limelight for a prolonged period are easily recollected compared to the
events kept in obscurity. The differences in ability to easily recall past events within dif-
ferent experiential context diminishes the reliability of risk management decision-making
under the availability heuristics.
The ease by which an assessor can search for real-life occurrence of similar past events
could also be a potential source of bias in intuitive risk management decision-making under
the availability heuristics. According to Tversky and Kahneman (1982a), most people find it
easy to imagine abstract occurrences in comparison to concrete events. Additionally, most
people find it easy to search for words by starting from the first letter, compared to the
middle or last letters. The implication is that the structure and composition of an event
may enhance or hinder the ability to easily recall a similar past occurrence which will in
effect reduce the reliability of predictions under the availability heuristics.
Another potential source of bias in intuitive prediction is the effect of anchor and adjust-
ment (Tversky and Kahneman 1982a). According to Chapman and Johnson (2002), esti-
mates are usually made from a starting ‘anchored’ value and adjusted to arrive at the final
estimate. The initial value may be given at the start of the estimate or deduced from com-
putations from the base information. Epley and Gilovich (2002) have suggested that adjust-
ments are normally insufficient, leading to errors in the final estimate. Kahneman (2011)
traces the cause of insufficient adjustment to the premature abrogation in the adjustment
process at the time when an assessor becomes unsure whether to continue adjusting. Kah-
neman (2011, p. 122) has further suggested that there are instances where the error in
estimation may not result from insufficient adjustment but rather the effect of perceiv-
ing the anchored value to be a case of suggestion, and thus stimulating associated evidence
which leads to the final estimate reflecting the prime characteristics of the anchored value.
The implication of the inherent biases of anchor and adjustment for judgements where the
validity of the base information, and reliability of the adjustment formulae are not verified
is that different individuals may likely achieve different judgements for the same event.
Research findings have also revealed that intuitive evaluation of compound events is
likely to result in biases and systematic errors (Cohen et al. 1972; Bar-Hillel 1973; Kahne-
man 2011; Appendix A). Bar-Hillel (1973) argues that probability estimation of conjunctive
events is usually overestimated whilst probability estimation of disjunctive events is usually
underestimated. In a study involving high school seniors and college students, participants
64 5 Construction Risk Management Decision-Making
were asked to bet on three classifications of simple, conjunctive, and disjunctive events,
involving the picking of coloured marbles from a parent population composed of coloured
and white marbles. The findings revealed a high preference for the simple and conjunc-
tive events compared to the preference for disjunctive event even though statistically the
disjunctive event presented the highest probability. Tversky and Kahneman (1982a) and
Kahneman (2011) attributed the lack of sensitivity between the probability values and pref-
erence to the error of anchor and adjustment associated with estimates of compound events.
The sum of probability estimates for compound conjunctive events are less than the proba-
bility estimates of each of the individual component events, whereas the sum of probability
estimates for compound disjunctive occurrences, is higher than the probability estimates
of the individual component events. In the process of estimating the probability value of a
compound event, the assessor normally starts with an initial value which is anchored on
the probability estimate of the individual component events and adjust them to achieve the
final estimate, which as revealed from the research findings, is normally closer to the initial
‘anchored’ value (Epley and Gilovich 2002).
The above analysis on the inherent biases in judgement and decision-making under intu-
ition in the light of the prevailing theoretical findings, which underline the over reliance
on intuition in construction risk management practices, probably explains why probabil-
ity predictions and quantification of construction risk events sometimes turn out different
from the actual (National Audit Office 2000; CBI 2010; Flyvbjerg et al. 2003).
risk aversion increases with an increase in the prospect of a loss. Kahneman (2011) further
suggests that in the process of making a choice, an individual evaluates the psychologi-
cal impact of the likely benefits and/or losses associated with the available options before
making a final choice. Most people are likely to become ‘risk takers’ and accept a negative
gambling offering a probability to lose, against an alternative option which offers certainty
of loss. Again, most people will likely become ‘risk averse’ and reject a mixed gambling
scenario offering the prospect for both gains and losses, against an alternative which offer
certainty of a gain.
A major limitation of prospect theory is that it mainly evaluates risk attitudes under
different risky scenarios without offering an explanation for the causes of differences in
risk attitude between different subjects under the same scenario. Slovic and Lichtenstein
(1983) have criticised prospect theory for failing to explain preference reversal, which is the
inconsistencies in an individual’s monetary evaluation of different preference options and
demonstration of risk preference. Slovic and Lichtenstein (1983) further explains that when
people are asked to choose between a prospect which offer a high probability of winning
a moderate monetary reward and an alternative prospect which offers a low probability of
winning a high monetary reward, most people will exhibit preferences which are inconsis-
tent with the tenets of the economic theories of preference by choosing the former prospect
whilst at the same time assigning a high monetary value to the latter prospect. Schoemaker
(1990) has also criticised prospect theory for not explaining the discrepancies in an indi-
vidual’s risk preference under domains of certainty and probability experienced in similar
risky scenarios.
Loosemore et al. (2006) also categorises risk perceptions within the construction project
delivery system into external stakeholders and project team members. This approach how-
ever appears simplistic. In as much as some theoretical contemporary approaches to risk
management advocates for inclusion of input from all project stakeholders (Slovic 2010),
research findings in risk management practices shows that final risk management analysis
and decision-making are made by project team members (Lyons and Skitmore 2004) and
not external stakeholders, which explains the variances in exposure, and resulting differ-
ences in risk perceptions between the two groups.
Analysis of the conceptual relationship between a ‘risk event’ and the potential ‘impact
on an objective’ (Loosemore et al. 2006), as discussed in Chapter 3, sheds light on a possible
source of risk perception categorisation, which addresses the deficiencies of the current
theories.
Drawing from the concept of systems decomposition (Carmichael 2006) which
was reviewed in Chapter 2, the perception component within the risk management
decision-making sub-system can be analysed as a lower hierarchy sub-subsystem as shown
in Figure 5.5.
The components within the perception sub-subsystem are what are defined as the differ-
ences in risk perceptions. The factors which causes the construction delivery system’s com-
ponents to differentiate and sub divide into additional groups (Walker 2007) as reviewed in
Chapter 2, also exposes the individuals within the subgroups to different kinds and degrees
of experiences. These experiences generate affective heuristics (Slovic et al. 2010) which
influences perception formation (Damasio 2006; Bateman et al. 2010; Slovic et al. 2010).
The differences in the objectives of the differentiated sub-systems (Carmichael 2006; Walker
66 5 Construction Risk Management Decision-Making
Environmental forces
Perception A
Perception C
Perception D
Interactions
Figure 5.5 Risk management perception sub-subsystems. Sources: Adapted from Carmichael
(2006) and Slovic (2010).
2015) of the construction delivery system also suggest that the impact of environmental
forces will result in differences in availability heuristics and therefore differences in per-
ceptions within the various sub-systems (Loosemore et al. 2006; Slovic et al. 2010). This
establishes a connection between the generating processes of systems differentiation and
the differences in perceptions. The differences in risk perceptions within the construction
project delivery system will therefore manifest along the pattern of the groupings of the
internal components: companies, specialist roles, and personal objectives (see Figure 5.6).
The analytical proposition has been described as differentiated risk perceptions.
The analytical proposition of differentiated risk perceptions appears consistent with the
tenets of culture theory which argues that social institutions and groupings are formed
based on common objectives, values, and perceptions (Thompson et al. 1990). It must how-
ever be emphasised that unlike culture theory which has been criticised for focusing just on
the differences between social institutions and groupings without offering explanation for
the complex differences in perceptions within their internal structures (Kamper 2000), the
underlying logic of differentiated risk perceptions, on the other hand, addresses this limita-
tion and further offers theoretical explanation for the generating processes of differentiation
in functionalities and perceptions of the internal components within social groupings.
Considering the previous discussions on biases and errors in judgment and decision-
making under intuition (Tversky and Kahneman 1982a,b,c; Kahneman and Tversky 1982b;
Kahneman 2011), which establishes the inherent systematic inefficiencies in intuitive
risk management systems coupled with the analytical implications of differentiated risk
perceptions and the theoretical evidence of high proportion of construction risk analysts
and decision-makers without formal training in risk management systems (Akintoye and
Macleod 1997; Kululanga and Kuotcha 2010), then the current intuitive construction risk
management decision-making systems (Akintoye and Macleod 1997; Lyons and Skitmore
2004; Kululanga and Kuotcha 2010) can be likened to a faulty vehicle being driven by an
untrained person who relies on instincts rather than structured procedure, the obvious
result will be a catastrophic accident. It is therefore not surprising that construction risk
management practices usually exhibit errors in the analysis and estimates (National Audit
Office 2000; CBI 2010) with minimal reliability in the risk responses (Flyvbjerg et al. 2003).
Construction delivery
system
Risk perception
subsystem
Figure 5.6 Systems differentiation generating differentiated risk perceptions. Sources: Adapted from Carmichael (2006), Loosemore et al. (2006), Walker
(2007, 2015), Slovic et al. (2010), and Kahneman (2011).
68 5 Construction Risk Management Decision-Making
5.6 Summary
Chapter 5 has applied the psychometric risk conceptual interpretation to examine the theo-
retical and practical issues of the current intuitive construction risk management practices.
The review has involved an evaluation of the two systems of thinking and decision-making
underpinning risk management systems. The quick approach is constantly active, utilis-
ing natural stimulus and reflexes to automatically process decision tasks. The effective-
ness of decision outcome under this approach depends on the rate of spontaneity. The
quicker the decision processing, the higher the quality of the decision outcome. The gradual
approach, on the other hand, requires deliberate activation through recollection of suitable
cognitive programme to analyse structured data. The decision-making requires alertness
as interferences disrupt the process. The quality of the decision outcome depends on the
decision-making setting and the level of attention received. Decision processing usually
commences in the quick approach by examining if there are previous experiences that com-
pares to the key attributes of the decision task and modelling the decision outcome in a
similar pattern. Where there are no similar previous experiences, the decision task is rolled
on to the gradual approach for rational analysis using an appropriate cognitive programme
(Kahneman 2011).
The remaining discussions focused on an in-depth analytical review of the quick
decision-making system utilised in the existing intuitive construction risk management
practices. Behavioural science theories confirm reliance on heuristics inspired personal
perceptions in quick decision processing. The key determinants of perceptions include
psychological, sociological, institutional, and cultural factors. The human brain stores
records of our everyday experiences and exposures in the form of perceptual images.
Attached to the records are stimulus relating to the type of emotion experienced at the time
of the exposure, termed affect. The perceptual images and associated feeling state are the
factors that guide the quick decision-making system. The presence of a mental record with
similar attributes to a given decision task will induce a feeling state like that experienced
at the time of the previous exposure, leading to the modelling of the decision outcome
in a similar pattern. A positive image of the likely impact of a future occurrence will be
classified as a positive risk to be enhanced, whilst a negative image will be interpreted as a
negative risk to be avoided, or managed.
There has been further analytical review of the impact of affect on the gradual
decision-making approach to ascertain if the low application of rationality in the existing
construction risk management systems is because of the influence of affective felling. The
theoretical evidence suggests both positive and negative impacts of affect on cognitive
reasoning. The established belief is affective feeling suppresses cognitive reasoning through
the inducement of physiological stimulus to take up the cognitive capacity in the human
brain, thereby giving a false sense of the quick decision-making approach being in control
of the situation, and therefore no need for considered cognitive analysis. There have also
been recent research investigations with consistent findings within different assessment
settings and demography, confirming a positive correlation between positive affective feel-
ing and cognitive reasoning. A comparative analysis of the established theory against the
contemporary empirical findings suggests indifference in the impact of affect on cognitive
5.6 Summary 69
reasoning. It is therefore possible that the low application of rational construction risk
management techniques and processes may not be related to the impact of affect.
The psychometric risk conceptual analysis extended into a review of construction
risk data presentation formats. The prevailing theory confirms application of qualitative
data including narratives, subjective statements, anecdotal scenarios, and quantitative data
expressed in the form of statistics and probability. Other studies have revealed risk data
presentation in the form of fuzzy linguistics variables. The historical development of the
risk terminology discussed in Chapter 3 confirmed the introduction of quantitative risk
data in the fourteenth century, starting from the northern Italian city states, then extending
to the other European nations, and subsequently to the multiple disciplines applying the
risk concept, including construction. The other theoretical evidence on the high incidence
of intuitive construction risk management practices in effect confirms intuitive processing
of quantitative risk data. The psychometric research findings however suggest that most
people do not understand statistical information and would rather prefer qualitative data
presentation formats. The theory also confirms that intuitive evaluation of statistics and
probability may be subject to biases and systematic errors. Other studies on the other hand
have revealed instances where experts have relied on their professional experiences to
achieve accurate intuitive evaluation. The low competency in scientific decision processing
techniques within the construction industry, however, suggests the prevailing intuitive
processing of quantitative risk data are being guided by heuristics, rather than expert
experiential knowledge.
The psychometric risk conceptual analysis has subsequently evaluated the systematic
biases and errors of intuitive decision-making under the representative, availability, and
anchor and adjustment heuristics. Intuitive decision processing under the representative
heuristics become susceptible to biases resulting from the misapprehension of the concept
of chance, and insensitivity to the impact of prior probability, sample size, reliability, and
validity of provided base information, on intuitive evaluation. Intuitive decision process-
ing under the availability heuristics also become susceptible to biases relating to how easily
past events and searches comparable to a given decision task can be retrieved. The errors of
intuitive decision-making under the anchor and adjustment heuristics emanates from the
misunderstanding of an anchored value to be identical to a final estimate, leading to the
premature abrogation of an adjustment process.
The establishment of the role of personal perceptions in intuitive risk management deci-
sion processing confirms the relevance of a robust risk perception categorisation. This will
facilitate the effective identification and harmonisation of the personal perceptions of the
project team members into the risk management process. The present theoretical explana-
tion for risk perception categorisation includes the differences in personality traits which
gives rise to ‘risk takers’, ‘risk averse’, and ‘risk makers’. The problem with this approach is
different studies using different research methodologies and settings have produced differ-
ent classifications for the same individual. Kahneman and Tversky’s prospect theory also
provides explanation for the differences in risk perceptions under different risky scenarios
(Kahneman and Tversky 1979). A major limitation of this approach is the lack of explana-
tion for the differences in risk attitude between different subjects under the same scenario.
Loosemore et al. (2006) in analysing construction stakeholders have also categorised the
sources of construction risk perceptions based on the external stakeholders and the project
70 5 Construction Risk Management Decision-Making
team members. This approach has been criticised for being too simplistic. Culture theory on
the other hand categorises risk perceptions based on the differences in common objectives
between social institutions and groups. This approach has also been criticised for failing to
explain the sources of the complex risk perception patterns within the internal structures
of social institutions. Analysis of the processes of systems decomposition and differentia-
tion discussed in Chapter 2, and the conceptual relationship between a ‘risk event’ and its
‘impact on an objective’ discussed in Chapter 3, has however revealed a novel theoretical
explanation termed, the differentiated risk perceptions. The empirical evaluation and valida-
tion of the novel theory and the wider implications for construction management research
and practice will be discussed in the case study chapters forming the Part 2 of the book.
71
The Part 1 chapters have employed sociological and psychometric risk concepts, to dis-
cuss the current construction risk management practices. Chapter 2 began the theoretical
review with an exposition of the UK construction industry, its contribution to the national
economy, and the policy aspirations aimed at operational and managerial efficiencies.
The construction industry was then examined using the principles of general system
theory. The operational and managerial processes at the corporate and project levels are
defined by their respective system’s objectives. The concept of systems decomposition has
enabled analytical review of the internal parts of the construction industry and processes, as
subsystems with different micro-objectives. The analysis of the project delivery processes
using the RIBA Plan of Works 2020 has revealed different stages aimed at the different
aspects of project delivery, with the output of a stage feeding into the input of the succeed-
ing stage. The analysis of the project management processes has also identified interactions
between different specialist roles possessing different micro-objectives and functionalities.
The construction industry being an open system suggests bilateral interactions between the
internal processes and the wider environment. The interactions with the multiple differen-
tiated functional environmental subsystems comprising political, economic, socio-cultural,
technological, ecological, and legal, influences the social construction of risk, in the context
of how risk is interpreted and treated.
Chapter 3 then provided a theoretical exposition on risk conceptualisation. Social threats
and uncertainties were conventionally defined and managed by the local stratified struc-
tures, with the religious and political setups at the top. The risk management strategy then
relied on morality, superstitions, taboos, and rituals. The replacement of the traditional
social order with contemporary multiple differentiated functional subsystems has expanded
the social discussion on risk into the public realm. The realisation of the ability of human
agency to alter the outcome of future events, rather than relying solely on the intervention
of religion and guided morality, has subsequently introduced risk communication into our
everyday activities.
The historical development of the risk concept confirms association with the emergence
of games of chance. Risk conceptualisation during the medieval period focused on
the physical and psychological threats associated with pirate attacks and other voyage
related calamities. Insurance policies and probability techniques were introduced in the
fourteenth century by merchants operating within the Italian city states, for shared risk
management. This was mainly in relation to the management of potential losses associated
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
72 Summary of the Part 1
with shipwreck, being the main risk of their trade. The risk terminology subsequently
expanded into the other European commercial industries, appearing in England in the
1830s, when it was applied in the insurance industry.
There has been further theoretical review of the different conceptual interpretations of
risk, including the realist, psychometric, sociological, edgework, and real and socially con-
structed perspectives. The psychometric risk viewpoint as applied in this book appreciates
the influence of perceptions and heuristics in risk analysis and the mediation of the struc-
tural discrepancies between the subjective and rational risk management approaches. The
sociological risk concept also as utilised in this book emphasises the influence of social
interactions and institutions in risk communication. The chapter concluded with an evalu-
ation of the risk concept as explained by Loosemore et al. (2006), which confirms theoretical
pre-requisition to establish the potential for a future occurrence to affect the objective or
interest of a subject, before assigning it as an associated risk event.
Chapter 4 expanded the risk discourse by examining the changing perspectives on
organisational strategies to risk management; from the traditional approach of relying on
insurances and contracts to transfer risk management responsibilities to third parties, to
contemporary practices of internal risk management systems. Taking a socially constructed
risk perspective, the principles of general system theory have been applied in modelling the
construction risk management process and the subsystems comprising, risk identification,
risk analysis, risk response, and risk review.
There has been further review of the construction risk management approaches com-
prising intuition and rationality. The intuitive approach utilises subjective methodologies
to process qualitative risk data. The rational approach on the other hand uses scientific
techniques to systematically process quantitative risk data. There have been other recent
research investigations involving the application of fuzzy logic in the processing of linguis-
tic variables, which has been argued as another form of the rational approach, with the
difference being the data presentation format (Arthur and Pryke 2014). The choice of a suit-
able risk management approach depends on the prime attributes of the risk management
decision setting. Decision settings associated with feelings are compatible with the intu-
itive approach, whereas decision tasks requiring structured cognitive analysis work well
with the rational approach. Behavioural science evidence confirms inherent limitations of
both approaches, which in effect suggest the benefit of combining principles of intuition
and rationality to achieve effective risk management systems. The current construction
management publications however confirm, high incidence of intuitive risk management
practices, by professionals with limited knowledge in scientific techniques. This has been
attributed to the lack of quantitative data to complement the structured decision process-
ing of the rational approach, low competency in scientific decision processing techniques,
and the comparatively low value of construction projects making it difficult to justify the
required financial investment in rational risk management systems. A major criticism of the
aforementioned empirical explanations has been the lack of adequate analytical evaluation
of the underlying behavioural influences. This has been addressed in Chapter 5.
Chapter 5 has applied psychometric risk conceptualisation in analysing the practical
and theoretical issues of the current intuitive construction risk management practices.
The discussion began with analytical review of the different systems of thinking and
decision-making. The quick system is always active and voluntarily producing perceptions,
Summary of the Part 1 73
impressions, insights, intentions, and feelings. The quality of the decision output is directly
related to the rate of spontaneity of the decision processing. The gradual decision-making
system however requires thoughtful activation of appropriate cognitive programme to
process structured quantitative data. The effectiveness of the gradual decision-making
system depends on the level of attention provided during the processing. The prevailing
behavioural science theories confirm constant interactions between the two systems.
Decision processing usually starts within the quick system, using heuristics to model
the potential outcome of a given decision task, in patterns that reflect a similar previous
occurrence. The absence of a related previous record results in the transfer of the decision
processing to the gradual system for analytical evaluation.
There has been further in-depth theoretical review of the quick decision-making systems
associated with the prevailing intuitive construction risk management practices. The psy-
chometric analysis of personal perceptions has revealed the key determinants to comprise
of psychological, sociological, institutional, and cultural factors. The empirical behavioural
science findings confirm that the human mind stores records of life experiences, in the form
of perceptual images and affective stimulus, which are recalled during intuitive risk man-
agement decision processing. The presence of a record comparable to the prime attributes
of the given decision task influences the structuring of the future event in an identical pat-
tern, and subsequent classification as either a positive or negative risk event. The theoretical
evidence also confirms indifference in the impact of affective heuristics on rational decision
processing. This in effect suggests a lack of association between the current low application
of rational techniques and the psychological impact of affect heuristics.
There has been further review of construction risk management data presentation
formats. Qualitative risk data include narratives, subjective statements, and anecdotal sce-
narios. Quantitative risk data are usually expressed in the form of statistics and probability.
Other studies have also identified the use of fuzzy logic in processing risk data defined
in linguistic variables. The previous discussions on the different systems of thinking and
decision-making confirm the relevance of facilitating complementarity between an applied
decision processing vehicle and the data presentation format. Qualitative data thrives with
quick decision processing, whereas quantitative data formats align with gradual decision
processing. The empirical evidence from construction management research however has
identified theoretically incompatible practices of intuitive processing of statistical and
probability data. This may partly have resulted from the historical development of the risk
terminology, as discussed in Chapter 3. The introduction of probability risk management
techniques in the fourteenth century, first within the Italian city states and then the
other European commercial industries have subsequently been extended to the wider
disciplines applying the risk terminology, including construction. The behavioural science
evidence notwithstanding confirms intuitive processing of statistics, and probability may
be susceptible to biases and systematic errors. There are some empirical studies that
have revealed the possibility to achieve accurate intuitive evaluation using professional
experiences. The minimal proficiency in scientific techniques within the construction
industry, nevertheless, confirms reliance of heuristics rather than expert knowledge in the
intuitive processing of statistical and probability risk data. The resulting systematic issue
has been the variances between most construction risk management estimates and the
actuals (National Audit Office 2000; Flyvbjerg et al. 2003; CBI 2010).
74 Summary of the Part 1
Part 2
Case Studies
77
The Part 2 chapters present findings from a case study empirical investigation, which anal-
yses the psychological and practical issues involved in the prevailing intuitive construction
risk management practices. The research theoretical propositions were developed from ana-
lytical review of the social construction of risk within the construction transformational
processes and the subjectivity of risk perceptions. The underlying theoretical themes, as dis-
cussed in the Part 1, includes general systems principles and processes (Checkland 1999;
Bertalanffy 1968, 2015; Cole 2004; Carmichael 2006; Walker 2007, 2015; Cole and Kelly
2020 – discussed in Chapter 2); the generation of construction project risk events from sys-
tems differentiation processes (Blanchard and Fabrycky 1998; Walker 2007, 2015; Slovic
et al. 2010 – discussed in Chapter 2); the different conceptual interpretations of risk (Zinn
2008 – discussed in Chapter 3); the construction risk management system (Smith et al.
2006; Dallas 2006; Walker 2007, 2015 – discussed in Chapter 4); decision-making systems
(Kahneman 2011 – discussed in Chapter 5); risk perception formation from affect heuris-
tics (Benthin et al. 1993; Slovic et al. 2010 – discussed in Chapter 5); subjectivity of risk
perceptions, and the inherent biases of intuitive decision making (Tversky and Kahneman
1982a,b,c; Kahneman and Tversky 1982a,b; Kahneman 2011 – discussed in Chapter 5); and
the generating processes of differentiated risk perceptions (Carmichael 2006; Walker 2007,
2015; Slovic et al. 2010; Kahneman 2011 – discussed in Chapter 5).
The Part 2 begins with Chapter 6 which describes the research methodology. The
research process has followed the five phases described by Denzin and Lincoln (2011).
The first phase defined the researcher’s philosophical tradition of constructivism ontology
and post-positivism epistemology. The next phase was the development of theoretical
paradigms and perspectives, compatible to the research philosophical stance. This involved
the conceptualisation of the research themes, using systems thinking and behavioural
science theories, and the application of a post-positivist research approach. The third
phase was the selection of the case study investigative strategy, to ensure methodological
compatibility. The fourth phase involved the selection of multiple data collection methods
in the form of interview, direct observation, and review of documentation and the applica-
tion of pattern matching and cross-case synthesis analytical methods to explore theoretical
replication. The last phase was the application of the four tests commonly used in social
science investigations including construct validity, internal validity, external validity, and
reliability, to confirm the quality of the empirical findings.
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
78 Overview of the Part 2
The research methodology is followed by the presentation of the case study empirical
data in Chapter 7. The presentation includes an outline of the key facts about the case study
projects, justification for the selection of the cases, criteria for the selection of the partici-
pants from each case, the applied data collection approaches, and an examination of the
case study findings based on the three research propositions and the 10 underling theo-
retical issues. Chapter 7 presents the full empirical data from the case study project 1 plus
summary of findings from the case study projects 2–4. The full empirical data from case
study projects 2–4 has been presented in Appendices B–D.
The Part 2 concludes with Chapter 8 which discusses the cross-case empirical findings.
The presentation has followed the 3 research propositions and 10 underlying theoretical
issues. The data analysis has applied the matching pattern and cross-case synthesis tech-
niques, to explore theoretical replication. The findings from research propositions 1 and
2 have been analysed using a novel matching pattern analytical tool termed differentiated
risk matrix. The findings from research proposition 3 on the other hand have been anal-
ysed using quantitative statistical tools. The data analysis has also involved an evaluation
of the wider analytical implications of the case study empirical data including emerging
theoretical insights and areas requiring further investigations.
The study’s limitations have been discussed in the context of the applied research
methodology and the scope of empirical investigations. The choice of the case study
research strategy exposes the empirical findings to the common criticisms associated with
methodological rigour and concerns on generalisation (Yin 2003, 2014). To address the
methodological concern, the study has followed the recommendations of Yin (2014), by
making use of multiple data collection approaches including interview, direct observation,
and review of documentation, to obtain empirical evidence from multiple sources, to
facilitate data triangulation and demonstration of construct validity. A case study protocol
defining the data gathering procedures, rules, and questions was designed prior to the
field data collection, to ensure standardisation of the empirical investigative process and
demonstration of reliability for the study findings. Lastly, the empirical evidence was
recorded in a case study data base, to ensure accurate transcription of the field observations
into the final case study report.
The generalisation concern has also been addressed by focusing the research discus-
sions on examining logical replication (Rowley 2002), to establish analytical generalisation,
rather than statistical generalisation, which is usually associated with quantitative research
strategies (Eisenhardt 1989; Yin 2014). It is therefore the intention of the researcher that the
study’s empirical findings are interpreted as analytical generalisations with the potential of
an identical future intuitive construction risk management study which applies the same
case study protocol, discovering findings consistent with the research theoretical proposi-
tions, rather than the statistical observations.
Another limitation of the study’s empirical investigative process is the difficulty in
demonstrating absolute intuitive decision processing for the interview settings. The diffi-
culty stems from the theoretical evidence of constant interactions between the intuitive
and rational approaches, to thinking and decision-making (Kahneman 2011), which makes
it difficult to logically tame the rational system in an interview setting. To address this
concern, the research protocol was deliberately designed to stimulate intuitive decision
processing, by allowing the study participants’ limited time to respond to the questions.
Overview of the Part 2 79
The time restriction was meant to impede activation of the rational system, and thereby,
limit the decision processing to the intuitive system which operates spontaneously and
effortlessly (Kahneman 2011). The collection of data from multiple sources using different
approaches has also facilitated data triangulation to demonstrate intuitive decision pro-
cessing, as evident in the similarities between the study participants’ risk identification on
hypothetical project settings and their availability and representative heuristics (theoretical
issue 3) and the direct observation of spontaneous responses for research propositions
1 and 2 and evidence of emotional discomfort during the deliberations for research
proposition 3. Again, it is the intention of the researcher that the study’s empirical findings
are interpreted as data obtained within a stimulated intuitive decision-making setting,
rather than an absolute intuitive decision-making environment devoid of rationality.
Lastly, it must be emphasised that the research empirical context is intuitive construction
risk management decision-making systems. It is therefore the intention of the researcher
that the study’s findings and conclusions are interpreted and applied within the framework
of the empirical context, rather than the general decision-making systems.
81
6.1 Introduction
Chapter 6 describes the research methodology utilised for the empirical investigation on
the practical and theoretical issues of the current intuitive construction risk management
practices. The research proposals have been developed from the analytical review of the
generating processes of the study’s novel differentiated risk perception model, expounded in
the Part 1. The research processes have also followed the five stages suggested by Denzin
and Lincoln (2011). The first stage defines the study’s philosophical position of construc-
tivism ontology and post-positivism epistemology. The subsequent stages centre on the
logical transition from the research study’s philosophical stance to theoretically compat-
ible research paradigm and perspective (Stage 2), research investigative strategy (Stage 3),
data gathering and analytical methods (Stage 4), and approach for evaluating the quality of
the research process and empirical findings.
analysis, lack of training in rational risk management techniques, the size and scope of
construction projects make it difficult to justify investment in rational risk techniques, and
the non-appreciation of the relevance of rational tools in construction risk management
practices (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha
2010). A major criticism on the validity and decisiveness of the above reasons identified
within the construction literature is that they appear to be merely documentation of
research responses, without an adequate psychological evaluation (see Table 4.2), despite
the fact that intuitive decision-making forms part of human behaviour pattern (Zajonc
1980; Forgas 2001).
Theoretical findings from the behavioural sciences, however, suggest that the sponta-
neous occurrence of perceptions is the prime driver behind intuitive risk management
practices, especially when we are confronted with complex decision scenario in the face
of limited scientific data (Slovic and Peters 2006; Bateman et al. 2010). Research findings by
March and Shapira (1987), Edwards and Bowen (1998), and Slovic et al. (2010) also suggest
that the limited application of rational risk management practices is influenced by the lack
of understanding and trust in rational tools and techniques.
Research findings within both the construction industry and psychology disciplines con-
firm the relevance of intuition in effective decision-making (Loosemore 2006; Slovic and
Peters 2006; Finucane et al. 2003). Kahneman and Tversky (1982b) have however suggested
that the principles of intuitive probability judgement utilising the representative and avail-
ability heuristics are not consistent with the laws of chance and statistics and are likely to
produce systematic errors.
Tversky and Kahneman (1982a) have suggested that the error in judgement under the
representative heuristics results from judging probability predictions by evaluating the
similarities between the provided sample information and pre-conceived stereotypical
characteristics of the parent population, without consideration to the prior probability
statistics (Kahneman and Tversky 1973) and insensitivity to sample size, which lead to
a false expectation of similarities in composition and characteristics, between sample
population and their parent population (Tversky and Kahneman 1982b; Kahneman 2011).
The other sources of error are misconception of the law of chance fuelled by an expectation
of balanced systematic repetition of past occurrences in future predictions (Tversky and
Kahneman 1982a); and insensitivity to the reliability and validity of the base information
for probability predictions (Tversky and Kahneman 1982a; Slovic et al. 1982). Tversky and
Kahneman (1982a) have further suggested the sources of error in judgement under the
availability heuristics to include the ease of retrieving past instances, which influence the
assignment of high probability predictions for events associated with past famous occur-
rences and where the assessor had a direct involvement, compared to regular occurrences
(Tversky and Kahneman 1982c; Slovic 1982); and the ability to search past occurrences
based on the structure and composition of the event (Tversky and Kahneman 1982a).
Another potential source of error in intuitive judgement is the effect of ‘anchor and adjust-
ment’ (Tversky and Kahneman 1982a; Kahneman 2011), where estimates are made from an
initial given or deduced ‘anchored’ value and adjusted to achieve the final value (Chapman
and Johnson 2002). According to Epley and Gilovich (2002), adjustments are normally
insufficient, resulting in errors in final estimates. Kahneman (2011) argues that insufficient
6.2 Research Proposal 83
adjustment results from premature abrogation in the adjustment process, and assessors per-
ceiving an ‘anchored value’ to be a case of suggestion. Other research findings suggest that
the effect of ‘anchor and adjustment’ accounts for errors in the probability evaluation of
compound events, in terms of, overestimation of conjunctive events, and underestimation
of disjunctive events (Cohen et al. 1972; Bar-Hillel 1973; Tversky and Kahneman 1982a;
Kahneman 2011; Appendix A).
Project failures within the construction industry may be partly attributed to the error of
anchoring and adjustments of compound events (Arthur and Pryke 2014). In the process of
estimating project cost and duration, the probability of achieving target completion dates
within budget for activities on the critical path are normally overestimated, giving false
hope that the project will achieve the target completion date and budget (National Audit
Office 2000; Flyvbjerg et al. 2003). The error arises when estimates are started from an ini-
tial value, anchored on the probability values of the individual component events, without
considering the fact that project success depends on the achievement of the target indica-
tors for all the activities on the critical path, not just a few. Thus, the impact of failure in
achieving the target indicators for a critical path activity on project success comprises the
direct impact from the specific activity, plus the indirect impact from the other critical path
activities, as triggered by externalities.
The error of ‘anchor and adjustment’ of compound events also leads to underestimation of
disjunctive floating risk events, giving an illusion of a low probability of occurring. Flyvbjerg
et al. (2003) in evaluating the causes of mega project failure in Europe, identified that some
of the risk events which accounted for the project failures were known at the project incep-
tion, but were never treated during the execution phase. A possible reason may be that the
risk evaluation was influenced by an initial low anchored probability value based on the
individual floating events, rather than a collective evaluation of the compound disjunctive
events.
Now concerning the implications of differentiated risk perceptions, the previous theoreti-
cal review on systems decomposition (Carmichael 2006 – carried out in Chapter 2) suggests
differences in functionalities of the specialist roles within the construction project delivery
system. The subsequent analytical review on the generating processes of systems differ-
entiation, and the sources of differentiation within the UK construction industry (Walker
2007, 2015 – reviewed in Chapter 2), coupled with the theoretical evidence on perception
formation from affect heuristics (Slovic et al. 2010 – reviewed in Chapter 5), and the concep-
tual relationship between a ‘risk event’ and its ‘impact on an objective’ (Loosemore et al.
2006 – reviewed in Chapter 3), suggest differences in impact and perceptions on poten-
tial risk events among the different specialist roles forming the construction project team
(Loosemore 2006). In a situation where the different specialist roles adopt different risk
management systems, there could be cross purpose strategies, which could be detrimental
to project success (Pryke and Smyth 2006; Cagno et al. 2007; Arthur and Pryke 2013). Even
in a situation where a joint risk management system is adopted, failure to recognise and
integrate the differing risk perceptions between the project team members could lead to
conflicts (Mullins 2005).
Again, defining risk perception categorisations along the lines of the differences in func-
tionalities of a system’s components suggests that the effectiveness of an intuitive construc-
tion risk management practice will depend on the scope of experiences and specialisation of
84 6 Research Proposal, Methodology, and Design
the assembled project team; individuals, specialist roles, and companies. Where the project
team lacks resources in a specific specialist role, the absence of an associated availability
heuristic (Kahneman 2011) will potentially affect the identification and management of
risk events within that specialist role (Benthin et al. 1993). Even within a multi-disciplined
project coalition, the exclusion of a specialist role at the project organisational level from
the risk management process is likely to diminish the effectiveness of the risk management
system, in identifying potential risk events associated with the specific past exposures of the
excluded specialist role (Bateman et al. 2010).
Lastly, the differences in micro-objectives of the components within differentiated
systems (Carmichael 2006; Walker 2007, 2015) suggest that the different decision-making
systems, reviewed in Chapter 5 will likewise exhibit differences in impact from systems
environmental forces. The implications being that, mixing tools and techniques from
different decision-making systems, as evident in the present construction risk manage-
ment practices of intuitive processing of statistical and probability data (Akintoye and
Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha 2010; Bowden et al.
2001, Lock 2003), may introduce conflicting responses which may impede the required
collaborative transformational processes of the construction delivery system (Walker 2007,
2015; Kahneman 2011).
The above analytical review on the implications of differentiated risk perceptions makes it
necessary to research into the following research question, and subsequently, three research
propositions:
Underlining Theories
1. Systems differentiation also generates differences in risk perceptions among the system
components – different specialist roles (Walker 2007, 2015; Kahneman 2011; Slovic
et al. 2010).
2. The differences in micro-objectives and functionalities of differentiated subsystems
(Carmichael 2006; Loosemore et al. 2006) suggest that, the impact of environmental
forces will result in differences in affective heuristics, and therefore differences in
perceptions within the system components – different specialist roles.
6.2.1.2 Proposition 2
Intuitive risk management practices under the representative and availability heuristics,
which exclude some of the project subgroupings, run the risk of failing to adequately iden-
tify potential risk events.
6.3 Research Philosophical Traditions, Axioms, and Methodology 85
Underlining Theories
3. Perceptions are formed from the representative and availability heuristics (Kahneman
2011; Slovic et al. 2010).
4. The inherent subjectivity of perceptions make it manipulative and less rational.
Additional information could change risk perception even when the base situation
has not changed – best case scenarios which lead to gaps between pre-commencement
estimates and actual values at completion stage (Flyvbjerg et al. 2003; Alhakami and
Slovic 1994; Slovic and Peters 2006).
5. An uncertainty becomes a risk event to a subject if its occurrence can potentially affect
the objective(s) of the subject (Loosemore et al. 2006), suggesting that where there is no
possibility of impact, the subject may fail to recognise the uncertainty as a risk event.
6.2.1.3 Proposition 3
Statistical and probability risk data presentation and analysis are less likely to be understood
and applied in intuitive risk management systems compared to the use of qualitative data
presentation formats and subjective assessment techniques.
Underlining Theories
6. Research findings have identified construction risk management to be dominated by
people without formal training in risk management systems using intuition and heuris-
tics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha
2010).
7. Psychometric research findings suggest that most people relate better to narratives,
models, illustrations, and anecdotal simulations better than statistical data (Slovic et al.
2010).
8. Most people do not trust, understand, or use statistics and probability estimates in their
risk management decision-making (March and Shapira 1987; Akintoye and Macleod
1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
9. Most people relate better to qualitative assessments than quantitative assessments
(Slovic et al. 2010).
10. Where a potential outcome of a future event evokes a strong affective memory, the asses-
sor may be insensitive to the probability predictions on the event (Slovic et al. 2010).
research question aimed at investigating the implications of the existing construction risk
management systems.
The subsequent literature review explored the psychological explanation for risk percep-
tion formation, the differences between the intuitive and rational decision-making systems,
and the theoretical explanation on the sources and structure of risk perceptions (Slovic et al.
2010; Bateman et al. 2010; Kahneman 2011). The criticisms against the prevailing theories
and concepts on risk perception categorisation, including differences in personality traits
(Weber and Milliman 1997; Smith et al. 2006; Chauvin et al. 2007), prospect theory (Kahne-
man and Tversky 1979; Kahneman 2011), differences between project team members and
external stakeholders (Loosemore et al. 2006), and culture theory (Thompson et al. 1990),
revealed the need for a novel theoretical explanation. Again, the psychological explana-
tion on the generation of risk perceptions from affect heuristics (Benthin et al. 1993; Slovic
et al. 2010) and the incompatibilities between the principles of intuitive decision process-
ing and rational analytical techniques (Tversky and Kahneman 1982a), in the light of the
theoretical evidence of intuitive processing of statistical and probability construction data
(Bowden et al. 2001; Lock 2003) and the quantitative processing of fuzzy linguistic vari-
ables (Byrne 1995; Cox 1999; Zeng et al. 2007; Khazaeni et al. 2012a,b) also revealed the
need for empirical investigations into the psychological concerns, and inherent systemic
errors of the present intuitive construction risk management practices, and subsequently,
the appropriate construction risk management data presentation format.
The final phase was the refinement of the initial broad question into the research question
described in Section 6.2; how does systems differentiation and affect heuristics impact on intu-
itive construction risk management decision-making systems. And in order to demonstrate
theoretical substantiation for the research question, the underlining analytical propositions
and theoretical issues were also defined.
Four cases comprising of two public sector and two private sector construction project
teams were selected, based on the following case study boundaries:
● Criteria 1: Construction sector boundary (housing project)
● Criteria 2: Procurement route boundary (design and build procurement)
● Criteria 3: Cultural and geographical boundary (project team members with working
experience in the UK construction industry)
● Criteria 4: Professional role boundary (project team members involved in design devel-
opment and risk management decision-making)
The research analytical strategy was aimed at exploring theoretical replication (Yin
2014), through examination of the empirical evidence against the three research propo-
sitions and ten underlining theoretical issues, for evidence of consistency. And to ensure
logical compatibility, the pattern matching, and cross-case synthesis analytical approaches
were applied (Yin 2014).
6.4 Summary
This chapter has developed the research design and methodology. The research proposals
were established through analytical review of the sources of risk perception categorisations,
resulting in the discovery of the novel differentiated risk perception model.
The research processes followed the five stages suggested by Denzin and Lincoln (2007,
2011). The first stage involved the definition of the study’s philosophical stance comprising,
constructivism ontology, and post-positivism epistemology. This was followed by the sec-
ond stage describing the use of comparable philosophical posture, in the interpretation of
the research phenomenon and strategies for data collection and analysis. This has been
reflected in the conceptualisation of risk as socially constructed and psychometric. The
third stage discussed the selection of the case study investigative method to promote theoret-
ical complementarity. The fourth stage further described the selection of multiple parallel
data gathering methods comprising interview, direct observation, and review of documen-
tation. The pattern matching and cross-case analytical approaches have also been utilised
to explore theoretical replications of the case study findings. The final stage employed the
four tests usually applied in social science research, including construct validity, internal
validity, external validity, and reliability to demonstrate the quality of the research design
processes and empirical findings.
The next Chapter 7 presents the empirical findings from the four case study projects.
89
Data Presentation
7.1 Introduction
This chapter presents the empirical evidence from the Cases. The chapter has been organ-
ised into six sections. Section 7.1 presents an overview of the chapter. Sections 7.2–7.5
present the empirical evidence from the cases 1–4. The presentation includes an outline
of the key facts about the respective case study projects and the rationale for their selec-
tion; a description of the case study participants and criteria for selection; the applied data
collection approaches; and finally, an examination of the case study findings based on the
3 research propositions and 10 underlining theoretical issues. The chapter concludes with
a summary of the data presentation chapter.
The empirical evidence was collected within four cases comprising two private sector
housing projects located in London, and two public sector housing projects located in
London and Chatham, Medway. Data collection was done mainly using semi-structured
interviews, with additional evidence collected through direct observations and review of
project documentation.
To ensure theoretical sampling, the case study participants were selected from the project
team members whose core duties involved design development and risk management
decision-making, including the project managers, contracts managers, technical managers,
architects, engineers, quantity surveyors, and clients. The emphasis on design development
is due to the close link between design failure and the other typical construction project
risk events. A failure to get the design principles right at the project inception as a result of
inadequate client brief, incorrect project specifications, or errors in design concepts often
results in later project changes with serious implications on project execution (Latham
1994; Egan 1998; Lock 2003; Flyvbjerg et al. 2003).
Analysis of the research findings have demonstrated substantial support for the 3 research
propositions and 10 underlying theoretical issues discussed in Section 6.2.
pool, gym, and sauna. The residential accommodation was split between 200 private for
sale units and 107 affordable housing. Table 7.1 provides key facts about the project.
The contracts manager who described himself as the project risk manager confirmed the
applied risk management protocol to be; individual trade subcontractors preparing their
separate risk registers which were passed on to the principal contractor to amalgamate into
the project risk register. The principal contractor’s core project team members then held
brainstorming meetings at the initial project delivery stages to review and update the project
risk register. The project risk events and management strategies predominantly from finan-
cial and health and safety perspectives were subsequently reported to the developer’s board
of directors as part of the project manager’s monthly report.
Case study project 1 was selected due to its project setting satisfying all the case study
boundary criteria.
allowed limited time in responding to the questions to facilitate restriction of their decision
processing within the intuitive system (Kahneman 2011). There have also been documen-
tary reviews from the project website for the key scheme details. Table 7.2 provides the key
facts about the case study participants.
Theoretical Issue 1
– Theoretical issue 1: Systems differentiation also generates differences in risk percep-
tions among the system components – different specialist roles (Walker 2007, 2015;
Kahneman 2011; Slovic et al. 2010).
The theoretical issue is to be confirmed if the empirical evidence reveals similarities in
the respondents’ risk perceptions on common construction project risk events and the risk
perception categories from Table 7.3 associated with their specialist background.
The related Interview Questions are 22 and 23 (see Appendix A). The rationale behind
these questions was to solicit the participants’ ranking of the typical construction project
risk events at the pre- and post-construction phases. The interview protocol was designed
to facilitate the generation of risk perceptions from the participants’ inherent heuristics,
at a time when they have not been introduced to any immediate project information. The
responses from these questions have therefore been termed grounded heuristics denoting
that they have been generated based on the research participants’ store of previous experi-
ences which over time have become grounded within their knowledge base.
Table 7.2 Key facts about case study 1 participants.
Contracts Technical Project Quantity
Specialist role manager manager manager surveyor Engineer Architect
Reference CM-1 TM-1 PM-1 QS-1 E-1 A-1
Years of involvement Over 5 yr Over 5 yr Over 5 yr Over 5 yr Over 5 yr Over 5 yr
in role
Years of involvement 37 yr 20 yr 20 yr 10 yr 14 yr 15 yr
in the UK
construction industry
Organisational Contractor Contractor Contractor Contractor Consultant Consultant
background
Years at current 3–5 yr 1–3 yr 1–3 yr 1–3 yr 1–3 yr 3–5 yr
organisation
Length of 1–3 yr 1–3 yr Less than 6 mo 1–3 yr 1–3 yr 1–3 yr
involvement in case
study project
Level of involvement Sometimes Sparsely Always Sometimes Sometimes Sometimes
in risk management
decision making
Level of training in On the job Continuing On the job CPDs, short duration CPDs and on the CPDs and on the
risk management informal professional informal formal course (less job information job information
systems experience developments (CPDs) experience than 6 mo) and on the experience experience
and on the job job information
information experience experience
7.2 Case Study Project 1 93
Interview Question 22 Drawing from your professional experiences what will you say are
the factors likely to affect project success (risk events) at the pre-construction phase (before
the start of actual construction on site)?
Findings: The responses have been presented in rankings corresponding to the order in
which the participants provided the risk events. A score of 5 has been assigned to the first
generated risk event, 3 for the second generated risk event, and 1 for the third generated risk
event. The risk events have also been categorised using the table of construction project risk
events (Table 7.3) developed from the cumulative responses of interview questions 22–24,
and structured along the lines of a similar previous classification by Perry and Hayes (1985).
The empirical findings have been summarised in Table 7.4.
The differentiated builder specialist sub-group comprising the contracts manager, tech-
nical manager, and project manager-generated risk perceptions in the following order of
ranking: ground condition related (first); construction related (second); logistics and com-
munication related (third). The designers comprising the engineer and architect also gen-
erated risk perceptions in the following order of ranking: ground condition related (first);
design and quality related (second); and construction related (third). The quantity sur-
veyor’s generated risk perceptions, on the other hand, were in the following order of rank-
ing: ground condition related (first); logistics and communication related (second); and
design and quality related (third).
In summary, the empirical findings have revealed evidence of differentiation within
the study participants’ risk perceptions on the typical pre-construction phase project risk
events. Apart from ground condition related risk events which appeared common in most
of the responses, the remaining generated risk perceptions exhibited reasonable correlation
with the study participants’ specialist background. The universality of ‘ground condition
related’ risk perceptions may have resulted from the dominant influence of an associated
construction industry affective heuristics component (Finucane et al. 2003; Slovic et al.
2010) within the study participants’ risk identification transformational systems (Walker
2015), leading to the generation of associated ‘ground condition related’ risk perceptions
(Benthin et al. 1993; Damasio 2006).
Interview Question 23 Drawing from your professional experiences, what will you say are
the factors likely to affect project success (risk events) at the construction phase (during
actual construction work on site)?
Findings: The responses have been presented in Table 7.5 using the same format as the
previous question Interview Question 22.
The differentiated builder specialist subgroup comprising the contracts manager, tech-
nical manager, and project manager generated risk perceptions in the following order of
ranking: ground condition related (first); construction related (second); legal and statutory
approval related (third). The risk perceptions generated by the designers comprising engi-
neer and architect were all design and quality related. The quantity surveyor’s generated
risk perceptions, on the other hand, were in the following order of ranking: construction
related (first); financial and commercial related (second).
7.2 Case Study Project 1 95
The empirical findings have once again revealed evidence of differentiation within the
study participants’ perceptions on the typical construction phase project risk events. The
correlation in patterns of the generated risk perceptions and the study participants’ special-
ist background, appears consistent with the research theoretical prediction of risk percep-
tion categorisation along the lines of the differences in specialist roles.
7.2 Case Study Project 1 97
Theoretical Issue 2
– Theoretical issue 2: The differences in objectives of differentiated subsystems
(Carmichael 2006; Loosemore et al. 2006) suggest that the impact of environmen-
tal forces will result in differences in affective heuristics, and therefore differences in
perceptions within the system components – different specialist roles.
Theoretical issue to be confirmed if the empirical evidence reveals similarities in the
respondents’ risk perceptions on hypothetical project settings and their specialist back-
ground.
The related Interview Question is 24 (see Appendix A). The rationale behind this question
was to solicit the participants’ risk perceptions on two hypothetical construction project
settings.
Interview Question 24 The pictures below depict hypothetical sites (edged in bold front)
being considered for residential development. The project is at the inception stage with the
following key project objectives:
– Design: 50 units to meet minimum building regulation requirements, lifetime homes,
secured by design, minimum 40% carbon reduction.
– Budget: Construction cost not to exceed £15 million.
– Programme: To be completed within two years.
Hypothetical site A: Brown field development within high-density urban area
Findings: The responses outline the participants’ risk perceptions on the hypothetical
projects. The risk events have been categorised using the table of construction project risk
events (Table 7.3). The findings have been summarised in Tables 7.6–7.8.
Table 7.8 compares the findings from the two hypothetical projects for evidence of con-
sistency in the responses. It also examines the pattern of the generated risk perceptions for
evidence of differentiation along the lines of the research participants’ respective specialist
background. The consistency analysis has further enabled filtration of the other extraneous
risk formation variables unrelated to specialist role affective heuristics.
The empirical findings have revealed evidence of differentiation within the generated risk
perceptions from the two hypothetical projects:
Case study 1
Cumulative Cumulative
Risk perception CM-1 TM-1 PM-1 for builders E-1 A-1 for designers QS-1
Force majeure related
Logistics and
communication related
Access and egress into Yes Yes Yes Yes Yes Yes
the site
Site setup and Yes Yes Yes
organisation
Financial and
commercial related
Completing within Yes Yes Yes
budget
(Continued)
100 7 Data Presentation
Case study 1
Cumulative Cumulative
Risk perception CM-1 TM-1 PM-1 for builders E-1 A-1 for designers QS-1
Legal and statutory
approval related
Achieving planning Yes Yes Yes Yes Yes
permission and
discharging conditions
Partywall issues Yes Yes Yes Yes
resulting from adjoining
properties
Political related
Construction related
Programming of Yes Yes Yes
construction works to
achieve completion date
Working at height Yes
Safety on site Yes Yes
Operational related
Evaluation of the evidence from the consistency analysis from the perspective of the
different specialist roles has also revealed patterns of categorisation consistent with the
differences in specialist background:
– The builders comprising the contracts manager, technical manager, and project manager
have exhibited risk perceptions associated with ground condition, design and quality,
legal and statutory approval, and construction related events. Ground condition-related
risk events were consistent in all the responses.
– The designers comprising the engineer and architect also displayed risk perceptions relat-
ing to ground condition, design and quality, and legal and statutory approval events.
Design and quality risk events were consistent in all the responses.
– The quantity surveyor, on the other hand, has exhibited risk perceptions relating to finan-
cial and commercial, and design and quality events.
7.2 Case Study Project 1 101
Case study 1
Cumulative Cumulative
Risk perception CM-1 TM-1 PM-1 for builders E-1 A-1 for designers QS-1
Force majeure related
Logistics and
communication related
Access to the site Yes Yes Yes
Financial and
commercial related
Procuring subcontractors Yes
and labour
Marketing of completed Yes Yes
properties
(Continued)
102 7 Data Presentation
Case study 1
Cumulative Cumulative
Risk perception CM-1 TM-1 PM-1 for builders E-1 A-1 for designers QS-1
Political related
Objection from local Yes
residence
Construction related
Programming Yes Yes Yes
construction works to
achieve project
completion date
Working at height Yes
Operational related
In summary, the empirical evidence from Theoretical Issues 1 and 2 have supported
research proposition 1. The realisation of patterns of risk perceptions correlating to the
differentiated specialist background of the study participants have confirmed the study’s
novel theoretical prediction of systems differentiation influencing risk perception categori-
sation within the construction project delivery system, with the structure reflecting the
differences in affective heuristics of the differentiated specialist roles. The consistent evi-
dence from the different construction delivery phases and hypothetical project settings has
further enhanced theoretical validity.
– 3. Perceptions are formed from the representative and availability heuristics (Kahneman
2011; Slovic et al. 2010).
– 4. The inherent subjectivity of perception makes it manipulative and less rational. Addi-
tional information could change risk perception even when the base situation has not
changed – best-case scenarios which lead to gaps between pre-commencement estimates
and actual values at completion stage (Flyvbjerg et al. 2003; Alhakami and Slovic 1994;
Slovic and Peters 2006).
– 5. An uncertainty becomes a risk event to a subject if its occurrence can potentially affect
the objective(s) of the subject (Loosemore et al. 2006), suggesting that where there is no
possibility of impact the subject may fail to recognise the uncertainty as a risk event.
Table 7.8 Consistency analysis of the risk perceptions from hypothetical project A and B (Case 1).
Case Study 1
Force Majeure
Related
Ground
Condition
Related
Contamination of Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
site
Structural
integrity/
considerations of
building
(Continued)
Table 7.8 (Continued)
Case Study 1
Achieving Yes
scheme design
standards
Civil Yes
infrastructure &
connections
Logistics &
Communication
Related
Site access Yes Yes Yes Yes Yes Yes Yes Yes Yes
Procurement of Yes
labour and
materials
Yes
Legal and
Statutory
Approval Related
Achieving Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
planning
permissions &
discharge of
conditions
Statutory Yes
approvals
including
building
regulations
(Continued)
Table 7.8 (Continued)
Case Study 1
Political Related
Construction
Related
Planning and Yes Yes Yes Yes Yes Yes Yes Yes
programming of
the construction
works
Operational
Related
7.2 Case Study Project 1 107
The presentation of findings has been organised under three headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants, supported the study’s theoretical predictions.
Theoretical Issue 3
– Theoretical issue 3: Perceptions are formed from the representative and availability
heuristics (Kahneman 2011; Slovic et al. 2010).
Theoretical issue to be confirmed if the participants’ risk identification on the hypothet-
ical projects correspond to their availability heuristics (evident from their responses on the
common construction project risk events).
The related interview questions are 22–24 (see Appendix A). The study data has been
examined for the evidence of the influence of grounded heuristics in the risk identifica-
tion processes, using hypothetical construction project settings. The interview protocol was
designed to first collect data on the participants’ grounded heuristics before asking them to
identify risk events from hypothetical project settings. The rational was to ascertain whether
the risk identification on the hypothetical project settings was done intuitively through
reliance on heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
grounded heuristics against their risk identification on hypothetical project settings. The
risk events have also been categorised using the table of construction project risk events
(Table 7.3). The findings have been summarised in Table 7.9.
The evidence from the empirical findings suggests substantial risk identification based
on the participants’ grounded heuristics, which confirms reliance on the representative and
availability heuristics:
– CM-1: Three out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Two out of the three risk identification from hypothetical
project B have corresponding grounded heuristics.
– TM-1: Two out of the three risk identification from hypothetical project A have corre-
sponding grounded heuristics. Three out of the five risk identification from hypothetical
project B have corresponding grounded heuristics.
– PM-1: Two out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Two out of the five risk identification from hypothetical
project B have corresponding grounded heuristics.
– QS-1: Four out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Two out of the three risk identification from hypothetical
project B have corresponding grounded heuristics.
– E-1: Two out of the three risk identification from hypothetical project A have correspond-
ing grounded heuristics. Two out of the two risk identification from hypothetical project
B have corresponding grounded heuristics.
– A-1: One out of the four risk identification from hypothetical project A has corresponding
grounded heuristics. Two out of the three risk identification from hypothetical project B
have corresponding grounded heuristics.
Table 7.9 Analysis of perception being formed from availability and representative heuristics (Case 1).
Case Study 1
CM-1 TM-1 PM-1 QS-1 E-1 A-1
Questions & Responses 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B
Logistics &
Communication Related
Site access Yes Yes Yes Yes Yes Yes Yes
Planning and programming Yes
of project delivery
Communication and Yes
information flow
Site setup, welfare and Yes Yes Yes
deliveries
No Yes Yes No No Yes No
Financial & Commercial
Related
Procurement of labour Yes Yes
and materials
Completing with budget Yes Yes
Marketing of completed Yes
properties
No Yes Yes No
Legal and Statutory
Approval Related
Achieving planning Yes Yes Yes Yes Yes Yes Yes Yes
permissions & discharge
of conditions
Third party issues- Yes Yes Yes Yes
partywall, etc
Statutory approvals Yes
including building
regulations
Yes No No No No No No
Political Related
Objection from local Yes
residence
No No No
Construction Related
Safety Yes Yes Yes Yes Yes Yes
Planning and Yes Yes Yes Yes Yes Yes Yes
programming of the
construction works
Managing the project team Yes
Appropriate subcontractor Yes
method statement
Yes Yes Yes Yes
Operational Related
3 of 4 2 of 3 2 of 3 3 of 5 2 of 4 2 of 5 4 of 4 2 of 3 2 of 2 2 of 2 1 of 4 2 of 3
112 7 Data Presentation
Theoretical Issue 4
– CM-1: Two out of the five risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Three out of the four risk assessment categories on
hypothetical project B were different from the initial assessment.
– TM-1: Three out of the five risk assessment categories on hypothetical project A were
different from the initial assessment. One out of the four risk assessment categories on
hypothetical project B was different from the initial assessment.
– PM-1: One out of the three risk assessment categories on hypothetical project A was
different from the initial assessment. One out of the four risk assessment categories on
hypothetical project B was different from the initial assessment.
Table 7.10 Analysis of the influence of additional information (exposed heuristics) on perception formation (Case 1).
Case Study 1
CM- 1 TM- 1 PM- 1 QS- 1 E-1 A-1
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Force Majeure Related
Fire destruction Yes
Flood destruction Yes
Landslips Yes
Yes Yes
Ground Condition Related
Contamination of site Yes Yes Yes Yes Yes Yes Yes Yes
Excavation of site Yes Yes
Polution from adjoining Yes Yes Yes
sites
Site constraints, eg Yes Yes
congested site
Ecological issues Yes Yes Yes Yes
Ground investigations Yes
Existing structures and Yes Yes
foundations
Buried services Yes Yes
Site Investigations Yes Yes Yes Yes
No No No No No Yes No No Yes No
(Continued)
Table 7.10 Analysis of the influence of additional information (exposed heuristics) on perception formation (Case 1).
Case Study 1
CM-1 TM-1 PM-1 QS-1 E-1 A-1
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Civil infrastructure & Yes
connections
Achieving Highways Yes Yes
Approval in Principle
(AIP) for civil works
New untested technology Yes
Design constraints/ Yes Yes Yes Yes
opportunities resulting
from surveys
Coordinating design with Yes Yes
construction method
Accuracy and Yes
appropriateness of design
specifications
Yes Yes Yes No Yes No No No
Logistics &
Communication Related
Site access Yes Yes Yes Yes Yes Yes
Site setup, welfare and Yes Yes Yes Yes
deliveries
Organising work into Yes
phases
Transportation of Yes Yes Yes
materials and construction
plants
Availability of specialist Yes Yes
labour
No No No No No Yes Yes
Financial & Commercial
Related
Procurement of labour and Yes
materials
Completing with budget Yes Yes
Availability of funds Yes Yes Yes Yes Yes
Marketing of completed Yes
properties
Project cash flow Yes Yes Yes Yes
Adequacy of insurance Yes Yes
Constructors and Yes Yes
supplier’s default
Exchange rate fluctuation Yes Yes
Inflation rate Yes Yes
Taxation rates Yes Yes
Yes Yes Yes No No No Yes
Legal and Statutory
Approval Related
Achieving planning Yes Yes Yes Yes Yes Yes Yes Yes
permissions & discharge of
conditions
Third party issues- Yes Yes Yes Yes Yes
partywall, etc
Statutory approvals Yes
including building
regulations
Personal liabilities Yes
No No No
(Continued)
Table 7.10 Analysis of the influence of additional information (exposed heuristics) on perception formation (Case 1).
Case Study 1
CM-1 TM-1 PM-1 QS-1 E-1 A-1
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Political Related
Objection from local Yes
residence
Changes in laws and Yes
regulations
Yes
Construction Related
Safety Yes Yes Yes Yes
Planning and Yes Yes Yes Yes
programming of the
construction works
Managing the project team Yes
Suitability of construction Yes Yes Yes Yes Yes Yes
method
Inclement weather Yes
No No Yes Yes No No Yes Yes Yes
Operational Related
Fluctuations in market Yes
demand
Ensuring fitness for Yes
purpose
Yes Yes
2 of 5 3 of 4 3 of 5 1 of 4 1 of 3 1 of 4 1 of 4 4 of 5 0 of 3 0 of 2 2 of 4 4 of 5
7.2 Case Study Project 1 117
– QS-1: One out of the four risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. Four out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– E-1: Zero out of the three risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. Zero out of the two risk assessment categories on
hypothetical project B was different from the initial assessment.
– A-1: Two out of the four risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Four out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
Theoretical Issue 5
– Theoretical issue 5: An uncertainty becomes a risk event to a subject if its occurrence can
potentially affect the objective(s) of the subject (Loosemore et al. 2006), suggesting that
where there is no possibility of impact the subject may fail to recognise the uncertainty
as a risk event.
Theoretical issue is to be confirmed if the participants fail to identify risk events outside
their specialist background (associated with lack of related grounded heuristics).
The related Interview Question is 24 (see Appendix A). The study data have been exam-
ined to see if the participants’ failure in identifying certain risk events can be attributed to
their lack of related grounded heuristics outside their specialist background. The interview
protocol was designed to solicit the participants’ risk perceptions on hypothetical projects
using their grounded heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
risk identification on the hypothetical project settings against the project cumulative risk
identification profiles to reveal the items missed by each participant. The risk events have
also been categorised using the table of construction project risk events (Table 7.3). The
findings have been summarised in Tables 7.11–7.13.
The data from the two hypothetical projects has been further analysed for consistency in
the participants’ missed risk events.
The empirical evidence and consistency examination have revealed that exclusion of a
specialist role from the risk identification exercise would have resulted in some of the risk
events being missed, with consequential impact on the robustness of the risk management
process and project performance. As an illustration, the responses from the hypothetical
project setting ‘A’ revealed that it was only the project engineer with his structural engi-
neering background who perceived ‘Agreement in Principle’ for structural works around
the existing roads to be a potential project risk. This implies that if the project engineer had
been excluded from the risk identification process, it would have affected the robustness
of the risk management processes, assuming the risk associated with the ‘Agreement in
Principle’ to be real.
Empirical evidence from the direct observations also revealed that the research partic-
ipants were generally calm and relaxed during the interrogations relating to Theoretical
Issues 3–5. Their responses were spontaneous and effortless suggesting intuitive decision
processing (Kahneman 2011).
Table 7.11 Effect of limited heuristics (excluding some project team members) on project risk identification – Hypothetical Project A.
Case study 1
CM-1 TM-1 PM-1 QS-1 E-1 A-1
Risk Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24A events events events events events events events events events events events events events
Force Majeure
Related
Ground Condition
Related
Demolition & Yes Yes Yes
excavation
Site contaminations Yes Yes Yes Yes
Site Yes Yes
constraints-
congested
Existing structures Yes Yes
and foundations
Buried services Yes Yes
Missed
Design & Quality
Related
Achievement of Yes Yes
scheme design
standards
Achieving Yes Yes
Highways Approval
in Principle (AIP)
for civil works
Availability and Yes Yes
feasibility of
services
connections
Missed Missed Missed
Logistics &
Communication
Related
Access and egress Yes Yes Yes Yes Yes
into the site
Site setup and Yes Yes Yes
organisation
Missed
(Continued)
Table 7.11 (Continued)
Construction
Related
Programming of Yes Yes Yes
construction works
to achieve
completion date
Working at height Yes Yes
Safety on site Yes Yes
Missed Missed Missed Missed
Operational
Related
Force Majeure
Related
Ground Condition
Related
Ground Yes Yes Yes Yes
contamination
Contamination Yes Yes
from adjoining
water bodies
Contamination Yes Yes
resulting from
previous landuse
Site investigations Yes Yes
Missed
(Continued)
Table 7.12 (Continued)
Case study 1
CM-1 TM-1 PM-1 QS-1 E-1 A-1
Risk Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24B events events events events events events events events events events events events events
Design & Quality
Related
Availability and Yes Yes Yes Yes Yes Yes
feasibility of
services
connections
Civil infrastructure Yes Yes
and connections
Missed
Logistics &
Communication
Related
Access to the site Yes Yes Yes
Missed Missed Missed Missed
Financial &
Commercial Related
Procuring Yes Yes
subcontractors and
labour
Marketing of Yes Yes
completed properties
Missed Missed Missed Missed
Legal & Statutory
Approval Related
Difficulty in Yes Yes Yes
achieving statutory
design standards
Achieving planning Yes Yes Yes Yes
permission &
discharging
conditions
Missed Missed
Political Related
Objection from Yes Yes
local residence
Missed Missed Missed Missed Missed
Construction
Related
Programming Yes Yes Yes
construction works
to achieve project
completion date
Working at height Yes Yes
Missed Missed Missed Missed
Operational
Related
Consistent
Missed risk missed
Reference Specialist role events – 24A Missed risk events – 24B risk event
CM-1 Contracts manager DQR, FCR DQR, LCR, FCR, LSAR FCR
TM-1 Technical manager DQR, FCR, CR PR, CR CR
PM-1 Project manager DQR, FCR FCR, PR FCR
QS-1 Quantity surveyor LSAR, CR GCR, LCR, PR, CR CR
E-1 Engineer LCR, FCR, CR LCR, FCR, LSAR, PR, CR LCR, FCR, CR
A-1 Architect GCR, CR LCR, FCR, PR, CR CR
GCR, ground condition related; DQR, design and quality related; LCR, logistics and communication
related; FCR, financial and communication related; PR, political related; LSAR, legal and statutory
approval related; CR, construction and site related.
In summary, the empirical evidence from the Theoretical Issues 3–5 has provided sub-
stantial support for research proposition 2. Analysis of the study participants’ risk eval-
uation on hypothetical construction projects against their grounded heuristics on typical
construction project risk events have confirmed intuitive risk management decision pro-
cessing guided by the representative and availability heuristics. All the study participants
apart from the project engineer who refused to re-assess his responses altered their risk
evaluation when they were exposed to additional information in the form of a schedule of
construction risk sources, which also confirms the susceptibility of perception to manip-
ulation. Further evaluation of the individual specialist roles’ risk identifications against
the cumulative risk events generated within the case have identified missed risk events
in the categories outside the participants’ respective specialist backgrounds. The impli-
cation being excluding a specialist role from the risk management decision-making sub-
system may affect robust risk identification, and subsequently, weaken the risk response
strategy.
– 7. Psychometric research findings suggest that most people relate better to narratives,
models, illustrations, and anecdotal simulations better than statistical data (Slovic et al.
2010).
– 8. Most people do not trust, understand, or use statistics and probability estimates in their
risk management decision-making (March and Shapira 1987; Akintoye and Macleod
1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
– 9. Most people relate better to qualitative assessments than quantitative assessments
(Slovic et al. 2010).
– 10. Where a potential outcome of a future event evokes a strong affective memory, the
assessor may be insensitive to the probability predictions on the event (Slovic et al. 2010).
The presentation of findings has been organised under five headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants have supported the study’s theoretical predictions on decision processing within the
construction risk management decision-making subsystem.
Theoretical Issue 6
– Theoretical issue 6: Research findings have identified construction risk management to
be dominated by people without formal training in risk management systems using intu-
ition and heuristics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga
and Kuotcha 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for intuitive
risk management approaches over rational tools and techniques.
The related Interview Questions are 7–11 (see Appendix A). The study data has been
examined for evidence on the level of knowledge and application of rational techniques
within the construction risk management decision-making subsystems. The interview pro-
tocol was designed to stimulate intuitive thinking and decision processing by allowing the
participants limited time to respond to the questions.
Findings: The findings from case study 1 have been summarised in Figures 7.1–7.9 and
Table 7.14.
Interview Question 7 Please indicate your level of education/training in project risk man-
agement tools and practices?
a. No formal education or training in project risk management (………)
b. On the job informal experience (………)
c. CPDs and seminars on project risk management (………)
d. Short duration formal course (less than six months) (………)
e. University diploma/degree in Project Risk Management – e.g. BSc Construction Risk
Management, MSc Risk Management (………)
Please specify title of degree/diploma…………………………………………………………
Findings: The responses have been presented in Figure 7.1.
126 7 Data Presentation
University diploma/degree 0%
Interview Question 8 Which of the following risk management techniques have you heard
of? (Please indicate with a tick)
Risk register ………… . . . .. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) …………… Risk breakdown analysis ………… . . . ..
Worse/best-case scenario ……………… Monte Carlo simulation/analysis ………… . . . .
Brainstorming………… . . . .. Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ……… . . . .. Other (please specify) …………………………
Findings: The responses against the risk management techniques have been presented in
Figure 7.2. Figure 7.3 also categorises the responses into intuitive and rational techniques.
Interview Question 9 Which of the following risk management techniques have you
applied on this project? (Please indicate with a tick)
Others 0%
Cultural beliefs – Intuitive 6.98%
Failure mode and effect analysis – Rational 0%
Monte Carlo simulation/analysis – Rational 4.65%
Risk breakdown analysis – Rational 4.65%
Personal feelings and values – Intuitive 14%
Probability matrix – Rational 6.98%
Gut feeling – Intuitive 11.63%
Fish bone – Rational 0%
Brainstorming – Intuitive 11.63%
Worse/best case scenario – Intuitive 11.63%
Personal experiences (heuristics) – Intuitive 13.95%
Fault tree – Rational 0%
Risk register – Intuitive 13.95%
Figure 7.2 Risk management techniques known by the participants (all responses).
Rational,
16.28%
Intuitive,
83.72%
Figure 7.3 Risk management techniques known by the participants (categorisation into intuitive
and rational).
Findings: The responses against the applied risk management techniques have been pre-
sented in Figure 7.4. Figure 7.5 categorises the responses into intuitive and rational tech-
niques.
Interview Question 10 On a scale of 1–3 please rank the top three (3) techniques which
you would prefer to use on your future risk management practices? You may select from the
following list: Risk register, Probability matrix, Fault tree, Personal feelings and values, Per-
sonal experiences (heuristics), Risk breakdown analysis, Worse/best case scenario, Monte Carlo
128 7 Data Presentation
Others 0%
Cultural beliefs – Intuitive 7.41%
Failure mode and effect analysis – Rational 0%
Monte Carlo simulation/analysis – Rational 0%
Risk breakdown analysis – Rational 3.70%
Personal feelings and values – Intuitive 14.81%
Probability matrix – Rational 0%
Gut feeling – Intuitive 14.81%
Fish bone – Rational 0%
Brainstorming – Intuitive 18.52%
Worse/best case scenario – Intuitive 3.70%
Personal experiences (heuristics) – Intuitive 18.52%
Fault tree – Rational 0%
Risk register – Intuitive 18.52%
Rational, 3.70%
Intuitive,
96.30%
Figure 7.5 Risk management techniques applied (categorisation into intuitive and rational).
simulation/analysis, Brainstorming, Failure mode and effect analysis, Fish bone, Cultural
believes, and Gut feeling (please note: A scale of 1 represents the highest preference, and a scale
of 3 represents the lowest preference).
1st ranked technique ………………………………………………………………………
2nd ranked technique ……………………………………………………………………
3rd ranked technique ……….………………………………………………………………
Findings: The responses against the preferred risk management techniques have been
presented in Figure 7.6. Figure 7.7 categorises the responses into intuitive and rational tech-
niques.
7.2 Case Study Project 1 129
Others 0%
Cultural beliefs – Intuitive 2.22%
Failure mode and effect analysis – Rational 0%
Monte Carlo simulation/ analysis – Rational 0%
Risk breakdown analysis – Rational 11.11%
Personal feelings and values – Intuitive 6.67%
Probability matrix – Rational 0%
Gut feeling – Intuitive 0.00%
Fish bone – Rational 0%
Brainstorming – Intuitive 20.00%
Rational,
11.11%
Intuitive,
88.89%
Figure 7.7 Risk management techniques preferred (categorisation into intuitive and rational).
A trend analysis of the responses on the risk management techniques known, applied,
and preferred has been carried out to see if there are any pattern in the responses that could
shed light on the rationale behind the present risk management practices. The evidence has
been presented in Figures 7.8 and 7.9.
The evidence from the trend analysis has revealed high responses for intuitive techniques
across the three assessment variables. The dominance of the intuitive decision processing
system in the existing construction risk management practices was very elaborate in the
responses, with all the specialist roles apart from the quantity surveyor confirming sole
130 7 Data Presentation
0
Others 0
0
Cultural beliefs – Intuitive 1 2
3
Failure mode and effect analysis – Rational 0
0
0
0
Monte Carlo simulation/analysis – Rational 0 2
5
Risk breakdown analysis – Rational 1
2
Personal feelings and values – Intuitive 3
4
6
0 Preferred
Probability matrix – Rational 0
3
0 Applied
Gut feeling – Intuitive 4
5
Fish bone – Rational 0 Known
0
0
Brainstorming – Intuitive 5 9
5
0
Worse/best case scenario – Intuitive 1 5
Personal experiences (heuristics) – Intuitive 5 13
6
Fault tree – Rational 0
0
0
Risk register – Intuitive 5 14
6
0 2 4 6 8 10 12 14 16
Figure 7.8 Trend analysis of the risk management techniques known, applied, and preferred.
120.00%
100.00%
96.30%
88.89%
80.00% 83.72%
60.00%
40.00%
20.00%
16.28%
11.11%
3.70%
0.00%
Known Applied Preferred
Intuitive 83.72% 96.30% 88.89%
Rational 16.28% 3.70% 11.11%
Figure 7.9 Trend analysis of the risk management techniques known, applied, and preferred
(categorisation into intuitive and rational).
reliance on intuitive practices. The project engineer with 14 years industry experience sug-
gested that the intuitive approach is the only option available to the construction industry.
A significant discovery from the trend analysis however is the variance between the per-
centages of rational techniques known, applied, and preferred. Out of the 16.28% ‘rational’
segment of the ‘known’ risk management techniques, only 22.7% are being applied on the
case study project, representing 3.7% of the responses on ‘applied’ risk management tech-
niques. There is however a 68.24% preference to apply the ‘known’ rational techniques
7.2 Case Study Project 1 131
Case study 1
Risk register Good for risk allocation and as an audit trail for checking risk 2
management responses
Enable all project risk to be documented in a single document 1
for monitoring and treatment
Simple way of recording and presenting project risk from 1
project inception
Personal Believe that is the only available medium for identifying and 1
experience treating project risk
Opportunity to draw from lessons learnt on previous projects 1
Brainstorming Effective in using your previous experience in risk identification 1
Opportunity to utilise knowledge from all the project team 2
rather than just one person
Personal feeling Opportunity to utilise the wealth of professional experiences in 2
and values project risk management
Risk breakdown Enable risk to be broken down for better clarity 1
analysis
Cultural belief An opportunity to transfer build-up experience into project risk 1
identification and treatment
132 7 Data Presentation
Theoretical Issue 7
– Theoretical issue 7: Psychometric research findings suggest that most people relate better
to narratives, models, illustrations, and anecdotal simulations better than statistical data
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for quali-
tative data presentation formats, over statistical data, and quantitative models.
The related Interview Questions are 12–14 (see Appendix A). The study data has been
examined for evidence on preference for construction risk management data presentation
format. The interview protocol was designed to stimulate intuitive thinking and decision
processing by allowing the participants limited time to respond to the questions.
Findings: The findings from case study 1 have been summarised in Figures 7.10–7.17.
Interview Question 12 Which of the following forms of risk data presentations do you
know? (Please tick all the applicable)
Narrative …… . . . .. Probabilities ……………… . . . .
Qualitative statements ………… Whole numbers ………… . . . .
Graphs ……… . . . .. Ratios ……………………………
Fractions ……… . . . . Scenarios …… . . . .
Simulations ……… . . . .. Other (please specify) ……………
Findings: The responses against the risk management data presentation formats have
been presented in Figure 7.10. Figure 7.11 also categorises the responses into qualitative
and quantitative presentation formats.
43.48%
56.52%
Qualitative Quantitative
Figure 7.11 Risk management data presentation formats known (categorisation into qualitative
and quantitative formats).
Interview Question 13 Which of the following forms of risk data presentations have you
applied on this project? (Please tick all the applicable)
Findings: The responses against the risk management data presentation formats have
been presented in Figure 7.12. Figure 7.13 also categorises the responses into qualitative
and quantitative presentation formats.
Interview Question 14 On a scale of 1–3 please rank the top three (3) risk data presentation
formats which you would prefer to use on your future risk management practices? You may
select from the following list: Narratives, Probabilities, Qualitative statements, Whole numbers,
Graphs, Ratios, Fractions, Scenarios, and Simulations (please note: A scale of 1 represents the
highest preference, and a scale of 3 represents the lowest preference).
Findings: The responses against the risk management data presentation formats have
been presented in Figure 7.14. Figure 7.15 also categorises the responses into qualitative
and quantitative presentation formats.
134 7 Data Presentation
81.82%
Qualitative Quantitative
A trend analysis of the responses on the risk management data presentation formats
known, applied, and preferred have been carried out to see if there are any pattern in the
responses that could shed light on the rationale behind the present risk management prac-
tices. The evidence has been presented in Figures 7.16 and 7.17.
The evidence from the trend analysis has revealed high responses for qualitative data pre-
sentation formats across the three assessment variables, which suggest prolonged existence
for the research participants’ present risk management practice of qualitative data process-
ing. This was strongly echoed by the technical manager who advised simplicity in construc-
tion risk data presentation to enhance application. A major concern from the empirical
evidence, however, is the disparity between the proportions of applied systems of risk man-
agement techniques presented under Theoretical Issue 6 versus the systems of applied data
7.2 Case Study Project 1 135
84.44%
Qualitative Quantitative
presentation formats. The empirical evidence under Theoretical Issue 6 has revealed exces-
sively higher dominance of intuitive risk management techniques (96.3%) suggesting the
need for corresponding higher application of qualitative data formats to ensure theoret-
ical complementarity in the decision processing system. The empirical evidence on the
applied data presentation formats, however, has exhibited comparatively lower dominance
for qualitative data presentation formats (81.82%), suggesting that some of the intuitive
decision processing techniques are being applied with theoretically incompatible quantita-
tive data presentation formats. The project manager’s admission of quantitative probability
data processing in the light of his other responses of sole reliance on intuitive techniques
(see Appendix D) is a clear confirmation of the theoretically incompatible decision process-
ing practice.
136 7 Data Presentation
Others
Figure 7.16 Trend analysis of the risk management data presentation formats known; applied, and
preferred.
90.00%
80.00%
77.55%
70.00% 70.59%
60.00% 60.71%
50.00%
40.00% 39.29%
30.00% 29.41%
20.00% 22.45%
10.00%
0.00%
Known Applied Preferred
Qualitative 60.71% 70.59% 77.55%
Quantitative 39.29% 29.41% 22.45%
Figure 7.17 Trend analysis of the risk management data presentation formats known, applied, and
preferred (categorisation into qualitative and quantitative).
Theoretical Issue 8
– Theoretical issue 8: Most people do not trust, understand, or use statistics and probability
estimates in their risk management decision-making (March and Shapira 1987; Akintoye
and Macleod 1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
7.2 Case Study Project 1 137
Theoretical issue to be confirmed if the empirical evidence reveals low level of knowledge
and application of statistics and probability.
The related Interview Questions are 15–18 (see Appendix A). The study data has been
examined for evidence on the participants’ level of knowledge and understanding of statis-
tics and probability assessment. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions intuitively without consulting any reference source.
Findings: The findings from case study 1 have been summarised in Table 7.15 and
Figures 7.18–7.21.
Interview Question 15 Which of the following statistical symbols are you familiar with?
Please indicate by selecting from the following list: summation, probability density
function, probability function, sample mean, population mean, median, variance, standard
deviation, sample standard deviation, and don’t know.
Symbol Name
P(A)
S
∑
A 1/7 30/35
B 97.9% 84%
C 1 : 9743 4:5
Case study 1
33.33%
66.67%
Interview Question 17 Which of the above risk events is certain not to occur?
Risk event A ……………………………… . . . ..
Risk event B ………………… . . . ..
Risk event C ……………………………………… . . . ..
All (risk events A, B, and C) ………………………
None of the risk events …………………… . . . ..
Don’t know ……………………………………… . . . ..
Findings: The percentage of correct responses has been presented in Figure 7.19.
66.67%
Interview Question 18 Which of the above risk events is certain to have the highest impact?
Risk event A ……………………………… . . . ..
Risk event B ………………… . . . ..
Risk event C ……………………………………… . . . ..
All (risk events A, B, and C) ………………………
None of the risk events …………………… . . . ..
Don’t know ……………………………………… . . . ..
Findings: The percentage of correct responses has been presented in Figure 7.20.
Figure 7.20 Risk event certain to 0.00%
have the highest impact.
100.00%
Wrong response
The empirical evidence has revealed significant low level of understanding of statistical
symbols and probability assessments among the case study participants:
– 7 out of 24 correct identification of statistical symbols which represents 29.17% (see
Table 7.15).
– 4 out of 18 correct responses for the probability assessments which represents 22.2%.
The four correct responses were provided by the quantity surveyor and engineer whose
specialist background tends to be mathematically oriented. The remaining participants
all gave wrong responses. Another significant empirical evidence is the erroneous linkage
of higher probability predictions to certainty of occurrence of an event, which may be due
to a general lack of understanding of the probability concept.
Theoretical Issue 9
– Theoretical issue 9: Most people relate better to qualitative assessments than quantitative
assessments (Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals high accuracy for risk
analysis expressed in whole numbers, qualitative statements, and scenarios, compared to
risk analysis expressed in percentages, decimals, and ratios.
The related Interview Questions are 19 and 20 (see Appendix A). The study data has
been examined for evidence on the participants’ level of knowledge and understanding of
qualitative and quantitative assessments. The interview protocol was designed to stimu-
late intuitive thinking and decision processing by allowing the participants limited time to
respond to the questions intuitively without using calculators.
140 7 Data Presentation
Findings: The findings from case study 1 have been summarised in Figures 7.21 and 7.22
and Table 7.16.
Interview Question 19 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. There is a 75.2% probability of half of the site being contaminated (………)
B. There is a 1 : 321 probability of the site being contaminated (………)
C. There is a 0.381 probability of the site being contaminated (………)
Findings: The percentage of correct responses has been presented in Figure 7.21.
83.33%
Interview Question 20 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. The probability of projects over running their initial budget is 3 out of 10 (………)
B. out of every projects overrun their initial budget (………)
C. After reviewing the financial records of the four completed construction projects, only
project A recorded a final cost higher than the initial budget estimate (………)
Findings: The percentage of correct responses has been presented in Figure 7.22.
100.00%
Correct response
7.2 Case Study Project 1 141
The empirical evidence has revealed higher percentage of accuracy for the qualitative
probability equations compared to the quantitative probability equations:
– Qualitative probability equation: Six out of six correct responses.
– Quantitative probability equation: One out of six correct response representing 16.67%.
Another significant evidence is the composition of the wrong responses. Four out of the
five wrong respondents selected the probability equation expressed in percentages, rep-
resenting 80% of the wrong responses. Further questioning on the rationale behind the
preference for the ‘percentage equation’ format revealed the following responses:
a. Percentages are what we are familiar with from our primary and secondary educa-
tion – three responses.
b. Percentages are what we use in our everyday working life – one response.
The above responses suggest that the participants who selected the percentage equation
were simply relying on their availability heuristics to intuitively generate the answers.
There were two further supplementary questions to solicit empirical evidence on the
participants’ experience with answering the two forms of probability equations. The find-
ings have been presented in Table 7.16. The empirical evidence has revealed the qualitative
equation format to be comparatively straightforward and easy to understand.
Table 7.16 Participants’ experience with answering the quantitative and qualitative probability
equations.
Theoretical Issue 10
– Theoretical issue 10: Where a potential outcome of a future event evokes a strong affec-
tive memory, the assessor may be insensitive to the probability predictions on the event
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals insensitivity to differ-
ent range of hypothetical probability predictions of asbestos contamination risk.
The related Interview Questions are 21i–21iii (see Appendix A). The study data has been
examined for evidence on the participants’ responses to the different range of hypothetical
asbestos contamination risk. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions.
Findings: The findings from case study 1 have been summarised in Table 7.17.
Table 7.17 The impact of probability predictions on events which evokes strong affective memories.
21i 95% probability of 99% probability of death Appoint a licenced contractor to carry 6 out of 6 (100%)
mesothelioma causing resulting from exposure out further surveys to identify and remove
asbestos fibre conta- the asbestos before the commencement of
mination on 100% of the site construction works
21ii 10% probability of 50% probability of asbestos Appoint a licenced contractor to carry 5 out of 6 (83.33%)
non-malignant asbestos infection out further surveys to identify and remove
contamination on 10% of the asbestos before the commencement of
the site construction works
Carry on construction works at the 1 out of 6 (16.67%)
non-contaminated area whilst carrying
out further investigations at the
remaining areas to identify and remove
any asbestos residue
21iii 1% probability of 1% probability of Appoint a licenced contractor to carry 5 out of 6 (83.33%)
non-malignant asbestos non-malignant pleural out further surveys to identify and remove
contamination within a infection the asbestos before the commencement of
specific 1% area of the site construction works
Carry on construction works at the 1 out of 6 (16.67%)
non-contaminated area whilst carrying
out further investigations at the
remaining areas to identify and remove
any asbestos residue
7.2 Case Study Project 1 143
Interview Questions 21 Asbestos related illness has been a major cause of construction
occupational death. According to the Health and Safety Executive in the United Kingdom,
there has been a recent surge in asbestos related illness and deaths (see the table and pic-
tures below).
Asbestos related deaths
1968 153
2009 2321 411
2010 2347 412
2011 2291 429
2012 2535 464
Source: Shutterstock.com
144 7 Data Presentation
21i Please kindly review the information above and complete the ‘Appropriate risk
response’ column of the ‘Hypothetical project risk evaluation matrix’ below. Please select
your response from any of the following options:
i. Appoint a licenced contractor to carry out further surveys to identify and remove the
asbestos before the commencement of construction works.
ii. Carry on construction works at the non-contaminated area whilst carrying out further
investigations at the remaining areas to identify and remove any asbestos residue.
iii. Supply protective masks to all operatives and carry on with the construction works.
iv. Carry on with the construction works and forget about any asbestos.
Hypothetical project risk evaluations matrix
Appropriate risk
Probability of occurrence Probability of impact response (please select
Project based on initial survey based on initial survey from i, ii, iii, iv above)
21ii What will be your risk response if the risk evaluation results were to change to
Appropriate risk
Probability of occurrence Probability of impact response (please select
Project based on initial survey based on initial survey from i, ii, iii, iv above)
(Please select your response from any of the options given under 21i.)
21iii What will be your risk response if the risk evaluation results were to change to
Appropriate risk
Probability of occurrence Probability of impact response (please select
Project based on initial survey based on initial survey from i, ii, iii, iv above)
(Please select your response from any of the options given under 21i.)
Findings: The participants’ risk responses under the different probability predictions have
been presented in Table 7.17.
The empirical evidence has revealed significant insensitivity to the different range of
hypothetical probability predictions on asbestos contamination. A total of 83.33% of the par-
ticipants consistently responded that they would appoint a licensed contractor to carry out
further asbestos surveys to identify and remove asbestos contaminated materials before they
commence site construction works. The different levels of probability risk predictions could
not alter their responses not even for the very low probability prediction of non-malignant
contamination within a small specified area. The project manager in particular responded
that he will take no chances with the lives of his workforce irrespective of the level of prob-
ability of asbestos risk.
The empirical evidence from the direct observations also revealed some level of emotional
discomfort for some of the participants during the interrogations especially the statistical
and probability assessments (Theoretical Issue 8) and the quantitative equations (Theo-
retical Issue 9). Some of the participants even inquired if they could search the statistical
symbols on the Internet and also process the quantitative equations with calculators. When
these were denied, the frustration became evident in their body language, confirming the
psychological conflict they were facing in attempting intuitive processing of quantitative
data formats.
146 7 Data Presentation
In summary, the empirical evidence from Theoretical Issues 6–10 has provided support
for research proposition 3. The findings have reviewed massive reliance on intuitive con-
struction risk management techniques. There has been further evidence of intuitive pro-
cessing of theoretically incompatible quantitative data formats especially quantitative prob-
ability by the study participants who have demonstrated low proficiency in statistical sym-
bols and probability judgement. The ensuing systematic issues evident from the empirical
findings are comparatively higher accuracy for the intuitive evaluation of qualitative risk
data formats over quantitative risk data formats and insensitivity to probability predictions
of emotive events.
– Research proposition 1: The empirical evidence from Theoretical Issues 1 and 2 have
supported research proposition 1. The findings on differentiation within the study par-
ticipants’ risk perceptions on the typical project risk events at the different construction
delivery phases, and different hypothetical project settings, have been very consistent
with the study’s theoretical predictions. The realisation of patterns of risk perceptions
correlating to the specialist background of the study participants, have subsequently
provided empirical support for the research novel proposition of, systems differentiation
influencing risk perception categorisation within the construction project delivery
system.
– Research proposition 2: The empirical evidence from Theoretical Issues 3–5 have pro-
vided substantial support for research proposition 2. The findings have revealed sub-
stantial correlation between the research participants’ risk perceptions on hypothetical
project settings, and their initial grounded heuristics on hypothetical construction project
risk events, which suggest intuitive decision processing based on the representative and
availability heuristics (Theoretical Issue 3). All the study participants altered their risk
evaluation, after they were exposed to additional assessment information, in the form of
Table 7.19 Key facts about case study 2 participants.
The project risk management protocol involved the preparation of a project risk register
by the principal contractor’s contracts manager, at the project inception. The principal con-
tractor, sub-consultants, and client then held periodic project team meetings at which point
the risk register was reviewed and updated.
Case study project 3 was selected due to its project setting satisfying all the case study
boundary criteria.
The individual responses confirmed sole reliance on intuitive risk management practices
for the project manager and contracts manager with the remaining differentiated special-
ist roles exhibiting comparative superior reliance on intuitive techniques (Theoretical
Issue 6). The findings have also revealed massive qualitative data application and pref-
erence. The application and preference of qualitative data formats was elaborate in all
the study participants’ responses. The project manager attributed the cause of his attrac-
tion to qualitative data to be the ease of recalling previous experiences to guide spon-
taneous risk assessments. Analysis of the empirical evidence has also revealed intuitive
processing of theoretically incompatible quantitative data formats especially quantitative
probability (Theoretical Issue 7). In addition, the study participants have demonstrated
low proficiency in statistical symbols and probability judgement. Apart from the project
manager, contracts manager, and quantity surveyor, the remaining differentiated spe-
cialist roles demonstrated a general lack of understanding of the probability concept
(Theoretical Issue 8). The ensuing systematic issues evident from the empirical find-
ings are: comparatively higher accuracy for the intuitive evaluation of qualitative risk
data formats over quantitative risk data formats (Theoretical Issue 9) and insensitivity to
probability predictions of emotive events (Theoretical Issue 10). In summary, the empir-
ical evidence has validated the psychological difficulties involved in mixing intuitive and
rational approaches and data presentation formats.
2011). There have also been documentary reviews from the project website for the key
scheme details. Table 7.23 provides the key facts about the case study participants.
7.6 Summary
This chapter has examined the extent to which the empirical evidence from the research
case studies reflected the three research theoretical propositions expounded in Section 6.2.
The empirical evidence was collected within four cases comprising two private sector
housing projects and two public sector housing projects using interview as the main data
gathering approach with additional evidence collected through direct observations and
review of project documentation.
The case study participants were selected from the project team members whose core
duties involved design development and risk management decision-making including the
project managers, contracts managers, technical managers, architects, engineers, quantity
surveyors, and clients.
The empirical evidence has revealed substantial support for the three research theoretical
propositions:
– Research proposition 1: The empirical findings from Theoretical Issues 1 and 2 have
revealed patterns of risk perceptions correlating with the specialist background of the
study participants. Apart from the ground condition-related risk events which appeared
common in the responses on the typical pre-construction project risk events, the remain-
ing generated risk perceptions under the different project delivery phases and hypothet-
ical project settings exhibited correlation with the study participants’ specialist back-
grounds.
– Research proposition 2: The empirical findings from Theoretical Issues 3–5 have also
provided substantial support for research proposition 2. The study participants’ risk
assessment on hypothetical project settings have produced findings correlating with
their grounded heuristics, which confirms intuitive decision processing inspired by the
representative and availability heuristics (Theoretical Issue 3). All the study participants,
except the project engineer from case 1, and the client development manager from
case 4 who declined to reappraise the hypothetical project settings, altered their initial
risk assessment profiles after being exposed to additional information in the form
of a schedule of construction risk sources. The variations occurred even though the
project settings remain the same, thereby validating the susceptibility of perceptions
to manipulation (Theoretical Issue 4). Further evaluation of the individual specialist
roles risk identifications against the cumulative risk events generated within the cases
have identified missed risk events in the categories outside the participants’ respective
specialist backgrounds. The implication being excluding a specialist role from the risk
management decision-making subsystem may affect robust risk identification and
subsequently weaken the risk response strategy (Theoretical Issue 5).
– Research proposition 3: The empirical findings from Theoretical Issues 6–10 have
similarly provided substantial support for research proposition 3. The empirical evi-
dence has revealed low-level training in formal risk management techniques and
intuitive-dominated risk management practices among the study participants (Theo-
retical Issue 6). There is also an evidence of disparities between the systems of applied
and preferred risk management techniques versus the applied and preferred risk data
presentation formats with evidence of intuitive processing of theoretically incompatible
7.6 Summary 159
quantitative probability data (Theoretical Issue 7). Most of the study participants have
exhibited low proficiency in statistical symbols and probability judgement (Theoretical
Issue 8). A review of the study participants’ risk assessment has further revealed com-
paratively higher accuracy for the intuitive evaluation of qualitative risk data formats
over quantitative risk data formats (Theoretical Issue 9) and insensitivity to probability
predictions of emotive events (Theoretical Issue 10).
The empirical evidence from the direct observations also revealed that the research par-
ticipants were generally calm and relaxed during the interrogations. Most of the responses
were spontaneous and effortless suggesting intuitive decision processing (Kahneman 2011).
Even when the participants were denied access to reference materials and calculators in
the course of responding to research proposition 3, the resultant physiological discomfort
exhibited by some of the participants could be argued as the psychological conflict they were
experiencing in attempting intuitive processing of quantitative data (Kahneman 2011).
The next Chapter 8 discusses the cross case empirical findings and their wider implica-
tions for construction risk management decision-making.
161
Application
8.1 Introduction
Chapter 8 discusses the cross-case research empirical evidence based on the three research
propositions developed in Section 6.2. The findings from the four case study projects and
their wider implications for construction risk management research and practice have been
analysed. The deliberations have involved comparative review of the behaviour patterns of
the different specialist roles, and the different risk management strategies, in the form of
internal risk management by the private sector cases and external risk management by the
public sector cases.
Before the discussions, it is important to emphasise the inherent limitation of the adopted
qualitative research approach. The empirical evidence has been collected from just four
case study projects, thereby restricting the analytical discussions to the testing of theoreti-
cal replications (Rowley 2002) to establish analytical generalisations rather than statistical
generalisations (Eisenhardt 1989; Yin 2014).
proposed by prospect theory (Kahneman and Tversky 1979; Kahneman 2011) has also been
criticised for failing to explain the factors which influence preference reversals (Slovic and
Lichtenstein 1983, p. 582) and the discrepancies in an individual’s risk preference under
domains of certainty and probability, experienced in similar risky scenarios (Schoemaker
1990). Loosemore et al. (2006) risk categorisation based on the differences between project
team members and external stakeholders also appears too simplistic, and falls short of
offering an explanation for the differences in perceptions within the project team and
external stakeholder subsystems. Lastly, risk perception categorisation along the lines of
the differences in objectives and values of social institutions as suggested by culture theory
(Thompson et al. 1990) has also been criticised for failing to explain the complex internal
structures of the institutions and subgroups, within the larger society (Kamper 2000).
The empirical investigation of research proposition 1 was centred on two underlining
theoretical issues:
● 1. Systems differentiation also generates differences in risk perceptions among the system
components – different specialist roles (Walker 2007, 2015; Kahneman 2011; Slovic et al.
2010).
● 2. The differences in micro-objectives and functionalities of differentiated subsystems
(Carmichael 2006; Loosemore et al. 2006) suggest that the impact of environmental forces
will result in differences in affective heuristics, and therefore differences in perceptions
within the system components – different specialist roles.
and compliance with statutory construction regulations, which appears consistent with
their specialist background.
The risk perceptions generated by the designer specialist role, made up of the archi-
tects and engineers from the four case study projects, centred on design and quality related
events, which also is consistent with their specialist background.
The quantity surveyor’s specialist role likewise generated risk perceptions consistent with
their financial, commercial, and logistical specialist services, on all the research cases except
case study 4, where the findings rather reflected the common project risk profile of design
and construction related risk events. The deviation on case study 4 may have arisen from the
interactions of the affective heuristics (Slovic et al. 2010) components within the quantity
surveyor’s risk perception formation system (Walker 2015). The previous theoretical discus-
sion on systems differentiation (Chapter 2 – Walker 2015) and the psychology of perception
formation (Chapter 5 – Slovic et al. 2010; Bateman et al. 2010; Kahneman 2011) confirms
that the components within the quantity surveyor’s risk perception formation system will
comprise heuristics acquired through multiple sources including personal, group, project,
organisational, and industrial background. The construction industry being an open sys-
tem (Cole 2004; Cole and Kelly 2020) also suggests both intra and inter interactions during
the processes of systems transformation. And according to Finucane et al. (2003), in a situ-
ation where there are multiple sources of stimuli feeding into a behaviour pattern, then the
dominant properties of the sources and the physiological state of the decision maker may
influence the ultimate behaviour pattern. In the light of the above, it can be hypothesised
that the dominant properties of the case study project 4 generated risks profile, coupled
with the physiological state of the quantity surveyor at the time of the interview may have
accounted for the suppression of the other affective heuristics components within the risk
perception formation system, leading to the generation of risk events reflecting the general
project risk profile, rather than the decision maker’s specialist background.
The clients’ dominant risk perceptions also reflected the common cross-case identified
ground condition related risk events. This may also have resulted from the dominant influ-
ence of an associated construction industry affective heuristics component within their risk
perception formation systems (Finucane et al. 2003; Slovic et al. 2010).
There appears to be no significant differences between the pattern of responses from the
private sector cases (case study projects 1 and 2) with their emphasis on internal risk man-
agement strategies and the public sector cases (case study projects 3 and 4), which practised
external risk management.
The design specialist role comprising the architects and engineers from the four case
study projects have also generated dominant design and quality related risk events, which
similarly is consistent with their specialist background.
The quantity surveyor specialist role also generated risk perceptions consistent with
their specialist services, with the dominant risk perceptions from financial and commercial
related events.
The risk perceptions generated by the clients centred mainly on construction related
events. This may have been influenced by the affective heuristics acquired from their pre-
vious working experiences as site manager (case study 3 client) and roofer (case study 4
client).
Once again, there appears to be no significant differences between the pattern of
responses from the private sector cases (case study projects 1 and 2) with their application
of internal risk management strategy and the public sector cases (case study projects 3 and
4) where the project risk management activities are outsourced.
developed for this research, called the differentiated matrix. The analysis involved the initial
scoring of the risk perceptions generated within the different specialist groups based on the
percentages of consistent responses realised within each risk category. This was followed by
a second consistency test of the risk perception category scores from the two hypothetical
project settings. The categories displaying consistency scores equal and above ‘partial’ were
further analysed for differentiation. The following assessment scales have been applied:
The assessment criteria were carefully selected after taking into consideration the range
in frequency of study participants within the individual specialist roles. The emphasis was
on achieving a balance between filtration of extraneous risk formation factors, and the
need to make allowance for incidental variability in affective heuristics. This was to ensure
correlation between the assessment scales and the proportion of participants offering the
responses, and ultimately restrict the differentiations to the risk perception categories dis-
playing consistent specialist role dominant responses across the two hypothetical project
settings. The classification of specialist role differentiated risk perceptions have therefore
been limited to the categories displaying differentiation scores above ‘substantial’. The final
assessment matrix has been detailed in Table 8.1.
The differentiated risk perception analysis has revealed the following pattern of
correlation:
The following are the pattern of correlation from the perspective of the specialist roles:
● Builders: Evidence of substantial correlation with ground condition and legal and statu-
tory approval related risk events. The initial examination of the empirical findings from
Table 8.1 Differentiated risk perception analysis (Cross case discussion).
Builders Designers Quantity Surveyor Client
Risk Category 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale
Force Majeuire 0.00% 9.00% -9% Substantial 4.50% Minimal 0.00% 0.00% 0.00% Total 0.00% None 0.00% 0.00% 0.00% Total 0.00% None 0.00% 50.00% -50.00% Partial 25.00% Minimal
Related
Ground 54.55% 90.90% -36.35% Partial 72.73% Substantial 37.50% 50.00% -12.50% Substantial 43.75% Partial 100.00% 50.00% 50.00% Partial 75.00% Substantial 0.00% 50.00% -50.00% Partial 25.00% Minimal
Condition
Related
Design & 45.45% 54.55% -9.10% Substantial 50.00% Partial 100.00% 100.00% 0.00% Total 100.00% Total 50.00% 75.00% -25.00% Substantial 62.50% Partial 100.00% 100.00% 0.00% Total 100.00% Total
Quality Related
Logistics & 72.73% 54.55% 18.18% Substantial 63.64% Partial 50.00% 25.00% 25.00% Substantial 37.50% Partial 100.00% 50.00% 50.00% Partial 75.00% Substantial 100.00% 50.00% 50.00% Partial 75.00% Substantial
Communication
Related
Financial & 36.36% 45.45% -9.09% Substantial 40.91% Partial 50.00% 25.00% 25.00% Substantial 37.50% Partial 100.00% 75.00% 25.00% Substantial 87.50% Substantial 0.00% 0.00% 0.00% None 0.00% None
Commercial
Related
Builders Designers Quantity Surveyor Client
Risk Category 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale
Legal & 100.00% 54.55% 45.45% Substantial 77.28% Substantial 75.00% 50.00% 25.00% Substantial 62.50% Substantial 50.00% 75.00% -25.00% Substantial 62.50% Partial 100.00% 50.00% 50.00% Partial 75.00% Substantial
Statutory
Approval
Related
Political 0.00% 9.00% -9.00% Substantial 4.50% Minimal 12.50% 0.00% 12.50% Substantial 6.25% Minimal 0.00% 0.00% 0.00% Total 0.00% None 0.00% 0.00% 0.00% Total 0.00% None
Related
Construction 54.55% 54.55% 0.00% Total 54.55% Partial 37.50% 25.00% 12.50% Substantial 31.25% Partial 25.00% 25.00% 0.00% Total 25.00% Minimal 0.00% 0.00% 0.00% Total 0.00% None
Related
Operational 0.00% 0.00% 0.00% Total 0.00% None 0.00% 0.00% 0.00% Total 0.00% None 0.00% 0.00% 0.00% None 0.00% None 100.00% 100.00% 0.00% Total 100.00% Total
Related
the two hypothetical project settings also revealed consistent correlation between the
project management role and construction related risk events.
● Designers: Evidence of total correlation with design and quality related risk events and
substantial correlation with legal and statutory approval related risk events.
● Quantity surveyor: Evidence of substantial correlation with financial and commercial,
logistics and communication, and ground condition related risk events.
● Client: Evidence of total correlation with operational and design and quality related risk
events; and substantial correlation with legal and statutory approval, and logistics and
communication related risk events.
In summary, the empirical evidence from theoretical issues 1 and 2 has revealed
substantial support for research proposition 1. Apart from ground condition related risk
being common in the responses on the typical pre-construction phase risk events, which
may have been precipitated by a corresponding dominant industrial experience affective
heuristics (Finucane et al. 2003; Slovic et al. 2010) and the incidental evidence of dominant
project experience affective heuristics in the case 4 quantity surveyor’s responses on typical
pre-construction phase risk events, the remaining empirical findings have demonstrated
correlation between the study participants’ specialist backgrounds and their generated risk
perceptions. The coagulation of identical risk perceptions within related specialist roles has
provided empirical substantiation for the research theoretical prediction of construction
risk perception categorisation reflecting the structure of the differentiated specialist roles.
The realisation of consistent empirical support across the different assessment settings
and cases have subsequently demonstrated theoretical replication for theoretical issues
1 and 2 and analytical generalisation for research proposition 1. Notwithstanding, the
incidental empirical observation of dominant affective heuristics from sources other than
specialist role experiences has also confirmed the ability of the other components within
the perception formation subsystem to influence risk perception categorisation within the
construction project delivery system. There is, however, the need for further empirical
investigations to test the sensitivity of affective heuristics acquired through the other
sources including personal, project, organisational, and industrial experiences.
The validation of research proposition 1 with consistent empirical observations under
different assessment methods, and cases where different risk management strategies were
applied, has inferably provided a novel robust theoretical philosophy for risk perception
categorisation which addresses the criticisms against the existing theories and concepts
on the subject including, differences in personality trait (Weber and Milliman 1997;
Smith et al. 2006), prospect theory (Kahneman and Tversky 1979; Kahneman 2011),
categorisation between the project team members and external stakeholders (Loosemore
et al. 2006), and culture theory (Thompson et al. 1990). The intuitive investigative context
of the empirical evidence, in the light of the previous theoretical review on the operational
variations between the different systems of thinking and decision-making (Kahneman
2011 – discussed in Chapter 5), however suggest that the validity of the research novel
philosophy may be limited to applications under intuitive decision-making.
The study’s novel philosophy of risk perception categorisation emanating from the
differences in specialist roles suggests the need for involving wide-range specialists in
intuitive construction risk management practices, to ensure robust risk identification
and treatment. The research theoretical review on the differences in micro-objectives
8.3 Research Proposition 2: Discussions 169
and functionalities of the differentiated specialist roles (Bertalanffy 1968; Blanchard and
Fabrycky 1998; Carmichael 2006 – reviewed in Chapter 2) coupled with the conceptual
relationship between a ‘risk event’ and its ‘impact on an objective’ (Loosemore et al.
2006 – reviewed in Chapter 3) suggests that there will be differences in risk interpretations
and actions from the different actors operating within the construction project delivery
system. In a situation where the different specialist roles decide to adopt different risk
management systems, there could be cross purpose strategies which could be damaging to
project success (Pryke and Smyth 2006; Cagno et al. 2007; Arthur and Pryke 2013). Even in
a situation where a joint risk management system is practised, the failure to recognise and
integrate the differences in affective heuristics of the assembled specialist roles could lead
to conflict (Mullins 2005). The validation of the research novel theoretical philosophy for
risk perception categorisation therefore suggests the need for expansion of the integrative
role of construction project management (Winch 2010; Walker 2015) to encompass both
the technical and affective heuristics differences of the different specialist roles, to foster
efficient risk management systems, and ultimately project success (Pryke and Smyth 2006;
Cagno et al. 2007).
heuristics, evident in their initial responses on the typical construction project risk events.
The theoretical issue was to be confirmed if the study participants risk identification on
hypothetical project settings exhibited similarities with their initial grounded heuristics.
The cross-case empirical findings have been presented in Table 8.2.
The empirical findings have revealed approximately 65% correlation between the study
participants’ risk perceptions on the hypothetical project settings and their previous
responses on the common construction industry risk events, generated through grounded
heuristics (theoretical issue 1). The strong association between the generated risk events
and the grounded heuristics suggests intuitive decision processing (Kahneman 2011).
The empirical evidence from the direct observations also revealed that the study partic-
ipants were generally calm and relaxed during the interrogations. Their responses were
spontaneous and effortless, further confirming intuitive decision processing of affective
heuristics (Benthin et al. 1993; Kahneman 2011).
The analysis of the cross-case empirical findings against the study participants’ years
of experience in the construction industry has revealed considerable correlation in rank-
ing between the average percentages of grounded heuristics inspired risk perceptions and
the average years of industrial experience within the cases. Case study 4 ranks the highest
in both the average years of industrial experience (23 years) and the average percentage
of grounded heuristics inspired risk perceptions (73.70%). Case study 3 likewise ranks the
lowest in both the average years of industrial experience (18 years) and the average per-
centage of grounded heuristics inspired risk perceptions (56.94%). The empirical evidence
appears consistent with the findings from a previous study which identified a positive corre-
lation between the extent of intuitive risk management practices, and the level of industrial
experience, for project managers (Maytorena et al. 2007). The correlation between years
of industrial experience and the degree of grounded heuristics influence in risk perception
formation may be attributed to the fact that progression in years of industry experience also
expands industrial exposure and storage of related affective stimuli (Slovic and Peters 2006),
which can easily be recalled during risk perception formation (Benthin et al. 1993). Anal-
ysis of the ranking based on the different construction sectors risk strategies did not reveal
any significant differences. The highest and lowest ranked cases were from the public sec-
tor with their external risk management strategy, whilst the middle ranks were occupied by
the private sector cases which practise internal risk management strategy. The absence of
significant pattern may be due to the triviality of management strategy affective stimuli in
determining the degree of influence of heuristics in risk perception formation. Further anal-
ysis of the rankings from the perspective of the different specialist roles also did not reveal
any significant correlation. This also suggests that the processes of risk perception forma-
tion from grounded heuristics may be indifferent to the nature of specialist role affective
stimuli.
Analysis of the cross-case empirical findings against the study participants’ level of train-
ing in construction risk management tools also did not reveal any significant pattern, even
for those with some form of rational risk management training. The absence of an inverse
relationship between the level of rational risk management training and the degree of influ-
ence of grounded heuristics in the risk perception formation suggests the suppression of the
rational decision-making system, and therefore restriction of the risk management process
within the intuitive decision-making system (Kahneman 2011).
Table 8.2 The influence of grounded heuristics on the risk identification process.
Specialist role
Specialist role Case 1 Case 2 Case 3 Case 4 average score
Project manager A: 2 out of 4 (50%) A: 3 out of 4 (75%) A: 3 out of 4 (75%) A: 3 out of 4 (75%) 69.59%
B: 2 out of 5 (40%) B: 4 out of 4 (100%) B: 3 out of 4 (75%) B: 2 out of 3 (66.7%)
Average: 45% Average: 87.5% Average: 75% Average: 70.85
Experience: 20 yr Experience: 16 yr Experience: 20 yr Experience: 21 yr 19.25
Training: 2 Training: 3 Training: 2, 3 Training: 2
Contracts manager A: 3 out of 4 (75%) A: 2 out of 3 (66.7%) A: 1 out of 2 (50%) A: 3 out of 4 (75%) 68.78%
B: 2 out of 3 (66.7%) B: 2 out of 3 (66.7%) B: 1 out of 2 (50%) B: 4 out of 4 (100%)
Average: 70.85% Average: 66.7% Average: 50% Average: 87.5%
Experience: 37 yr Experience: 36 yr Experience: 16 yr Experience: 10 yr 24.75
Training: 2 Training: 2 Training: 3 Training: 2, 3
Technical manager A: 2 out of 3 (66.7%) A: 2 out of 5 (40%) A: 3 out of 3 (100%) 69.45%
B: 3 out of 5 (60%) B: 1 out of 2 (50%) B: 4 out of 4 (100%)
Average: 63.3% Average: 45% Average: 100%
Experience: 20 yr Experience: 15 yr Experience: 19 yr 18
Training: 2, 3 Training: 2, 3 Training: 2, 3, 4
Architect A: 1 out of 4 (25%) A: 1 out of 4 (25%) A: 1 out of 3 (33.3%) A: 4 out of 5 (80%) 45.41%
B: 2 out of 3 (66.7) B: 1 out of 3 (33.3%) B: 1 out of 2 (50%) B: 3 out of 6 (50%)
Average: 45.85% Average: 29.15% Average: 41.65% Average: 65%
Experience: 15 yr Experience: 6 yr Experience: 12 yr Experience: 18 yr 12.75
Training: 2, 3 Training: 2 Training: 2 Training: 2, 3
(Continued)
Table 8.2 (Continued)
Specialist role
Specialist role Case 1 Case 2 Case 3 Case 4 average score
Engineer A: 2 out of 3 (66.7%) A: 3 out of 4 (75%) A: 1 out of 3 (33.3%) A: 3 out of 3 (100%) 84.38%
B: 2 out of 2 (100%) B: 2 out of 2 (100%) B: 1 out of 1 (100%) B: 2 out of 2 (100%)
Average: 83.35% Average: 87.5% Average: 66.65% Average: 100%
Experience: 14 yr Experience: 7 yr Experience: 19 yr Experience: 20 yr 15
Training: 2, 3 Training: 2, 3 Training: 2 Training: 2
Quantity surveyor A: 4 out of 4 (100%) A: 2 out of 4 (50%) A: 3 out of 4 (75%) A: 3 out of 5 (60%) 67.93%
B: 2 out of 3 (66.7%) B: 2 out of 4 (50%) B: 2 out of 3 (66.7%) B: 3 out of 4 (75%)
Average: 83.35% Average: 50% Average: 70.85% Average: 67.5%
Experience: 10 yr Experience: 36 yr Experience: 25 yr Experience: 30 yr 25.26
Training: 2, 3, 4 Training: 2, 3 Training: 3 Training: 2
Client A: 2 out of 4 (50%) A: 1 out of 4 (25%) 31.25%
B: 1 out of 4 (25%) B: 1 out of 4 (25%)
Average: 37.5% Average: 25%
Experience: 16 yr Experience: 46 yr 31
Training: 2 Training: 2
Case average score 65.28% 60.98% 56.94% 73.70% 65%
19 19 18 23 20
Specialist role
Specialist role Case 1 Case 2 Case 3 Case 4 average score (%)
Project manager A: 1 out of 3 (33.3%) A: 2 out of 5 (40%) A: 3 out of 5 (60%) A: 3 out of 5 (60%) 45.41
B: 1 out of 4 (25%) B: 2 out of 5 (40%) B: 1 out of 4 (25%) B: 4 out of 5 (80%)
Average: 29.15% Average: 40% Average: 42.5% Average: 70
Contracts manager A: 2 out of 5 (40%) A: 2 out of 5 (40%) A: 2 out of 4 (50%) A: 1 out of 4 (25%) 48.75
B: 3 out of 4 (75%) B: 3 out of 5 (60%) B: 3 out of 4 (75%) B: 1 out of 4 (25%)
Average: 57.5% Average: 50% Average: 62.5% Average: 25%
Technical manager A: 3 out of 5 (60%) A: 1 out of 4 (25%) A: 2 out of 4 (50%) 59.43
B: 1 out of 4 (25%) B: 4 out of 5 (80%) B: 3 out of 4 (75%)
Average: 63.3% Average: 52.5% Average: 62.5%
Architect A: 2 out of 4 (50%) A: 1 out of (33.3%) A: 3 out of 5 (60%) A: 1 out of 5 (20%) 51.25
B: 4 out of 5 (80) B: 2 out of 3 (66.7%) B: 3 out of 5 (60%) B: 2 out of 5 (40%)
Average: 65% Average: 50% Average: 60% Average: 30%
Engineer A: 0 out of 3 (0%) A: 2 out of 4 (50%) A: 2 out of 3 (66.7%) A: 2 out of 3 (66.7%) 63.9a)
B: 0 out of 2 (0%) B: 1 out of 2 (50%) B: 3 out of 4 (75%) B: 3 out of 4 (75%)
Average: 0% Average: 50% Average: 70.85% Average: 70.85%
Quantity surveyor A: 1 out of 4 (25%) A: 1 out of 4 (25%) A: 2 out of 4 (50%) A: 1 out of 5 (20%) 40
B: 4 out of 5 (80%) B: 2 out of 5 (40%) B: 2 out of 5 (40%) B: 2 out of 5 (40%)
Average: 52.5% Average: 32.5% Average: 45% Average: 30%
Client A: 2 out of 4 (50%) A: 0 out of 4 (0%) 45a)
B: 2 out of 5 (40%) B: 0 out of 4 (0%)
Average: 45% Average: 0%
Case average score 53.49%a) 45.83% 54.31% 48.06%a) 50a)
Project manager A: 2 A: 2 A: 3 A: 4
B: 2 B: 2 B: 4 B: 4
Missed: FCR Missed: LCR Missed: Missed: CR
FCR, OR
Contracts manager A: 2 A: 3 A: 5 A: 4
B: 4 B: 2 B: 5 B: 3
Missed: FCR Missed: None Missed: GCR, Missed: CR, OR
DQR, CR, OR
Technical manager A: 3 A: 2 A: 5
B: 2 B: 4 B: 3
Missed: CR Missed: DQR Missed: LCR,
FCR, OR
Architect A: 2 A: 3 A: 4 A: 3
B: 4 B: 3 B: 6 B: 1
Missed: CR Missed: Missed: GCR, Missed: OR
FCR, CR LCR, FCR, OR
Engineer A: 3 A: 2 A: 5 A: 5
B: 5 B: 4 B: 7 B: 5
Missed: LCR, Missed: Missed: GCR, Missed: GCR,
FCR, CR LCR, CR LSAR, CR, OR FCR, OR
Quantity surveyor A: 2 A: 2 A: 3 A: 3
B: 4 B: 2 B: 5 B: 3
Missed: CR Missed: CR Missed: DQR, Missed: CR, OR
LSAR, OR
Client A: 3 A: 4
B: 4 B: 3
Missed: Missed: GCR,
FCR, CR FCR, CR
GCR, ground condition related; DQR, design and quality related; LCR, logistics and communication
related; FCR, financial and communication related; LSAR, legal and statutory approval related; CR,
construction related; OR, operational related.
specialist groups based on the percentages of consistent missed events realised within each
risk category. This was followed by a second consistency test of the missed risk percep-
tion category scores from the two hypothetical project settings. The categories displaying
consistency scores equal and above ‘partial’ were further analysed for differentiation. The
following assessment scales have been applied:
The assessment criteria as previously explained were carefully selected after taking into
consideration the range in frequency of the study participants within the individual special-
ist roles. The emphasis was on achieving a balance between filtration of extraneous vari-
ables and the need to make allowance for incidental variability in affective heuristics. This
was to ensure correlation between the assessment scales and the proportion of participants
who missed the risk events, and ultimately restrict the differentiations to the missed risk
perception categories displaying consistent specialist role dominance across the two hypo-
thetical project settings. The classification of specialist role differentiated missed risk per-
ceptions that have therefore been limited to the categories displaying differentiation scores
above ‘substantial’. The final assessment matrix has been detailed in Table 8.5. Table 8.6 is a
comparison of the missed risk assessment against the evidence from the differentiated risk
perceptions carried out under theoretical issue 2.
The differentiated missed risk perception analysis has revealed missed specialist-role
events outside their related specialist backgrounds. The inverse correlation between the
specialist-role missed events and their related differentiated risk perceptions evident from
Table 8.6, further provides substantiation on the psychological difficulties in intuitive risk
identification of events outside our sphere of heuristics (Loosemore et al. 2006; Slovic 2010;
Kahneman 2011).
In summary, the empirical findings from theoretical issues 3–5 have revealed substantial
support for research proposition 2. The research participants’ failure in identifying risk
events outside their specialist backgrounds has provided empirical substantiation for the
research theoretical prediction that intuitive risk management practices which exclude
some of the project team members, run the risk of failing to adequately identify project
risk events, especially those associated with the specialist background of the excluded
members. The realisation of consistent empirical support across the different specialist
roles and risk management strategies from the four cases has confirmed theoretical
replication of the study theoretical predictions, and subsequently analytical generalisation
for research proposition 2.
The empirical evidence on the psychological difficulties involved in the intuitive risk
identification of events outside the scope of a specialist background, coupled with the other
empirical evidence validating the research’s novel theoretical explanation of risk perception
categorisation along the lines of the differences in specialist affective heuristics suggests
that the effectiveness of an intuitive risk management system will depend on the scope
of professional expertise and experiences of the assembled project team within the con-
struction delivery system. Where the project team lacks resources in a specific specialist
role affective heuristics, the absence of an associated availability and representative heuris-
tics (Tversky and Kahneman 1982a,b,c; Kahneman and Tversky 1982a,b; Kahneman 2011)
may potentially constrain the identification and management of related risk events (Ben-
thin et al. 1993; Damasio 2006). Even within a multi-disciplined project team, the exclusion
of a specialist role from the risk management process may likewise weaken the robustness
of the risk identification and treatment processes, especially for the risk events relating to
the excluded specialist role affective heuristics. The empirical findings therefore suggest the
need for careful selection of project team members with the appropriate grounded heuris-
tics to ensure effective risk management systems (Arthur 2018). The formation of project
Table 8.5 Differentiated missed risk perception analysis (Cross-case discussions).
Builders Designers Quantity Surveyor Client
Risk Category 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale
Force Majeure 100.00% 90.90% 9% Substantial 95.45% Substantial 100.00% 100.00% 0.00% Total 100.00% Total 100.00% 100.00% 0.00% None 100.00% Total 100.00% 50.00% 50.00% Partial 75.00% Substantial
Related
Ground 45.45% 9.00% 36.45% Partial 27.23% Minimal 62.50% 50.00% 12.50% Substantial 56.25% Partial 0.00% 50.00% -50.00% Partial 25.00% Minimal 100.00% 50.00% 50.00% Partial 75.00% Substantial
Condition
Related
Design & 54.54% 45.45% 9.09% Substantial 50.00% Partial 0.00% 0.00% 0.00% Total 0.00% None 50.00% 25.00% 25.00% Substantial 37.50% Partial 0.00% 0.00% 0.00% Total 0.00% None
Quality Related
Logistics & 27.27% 45.45% -18.18% Substantial 36.36% Partial 50.00% 75.00% -25.00% Substantial 62.50% Partial 0.00% 50.00% -50.00% Partial 25.00% Minimal 0.00% 50.00% -50.00% Partial 25.00% Minimal
Communication
Related
Financial & 63.63% 54.55% 9.08% Substantial 59.09% Partial 50.00% 75.00% -25.00% Substantial 62.50% Partial 0.00% 25.00% -25.00% Substantial 12.50% Minimal 100.00% 100.00% 0.00% Total 100.00% Total
Commercial
Related
(Continued)
Table 8.5 (Continued)
Builders Designers Quantity Surveyor Client
Risk Category 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale 24A 24B Variance Scale Average Scale
Legal & 0.00% 45.45% -45.45% Partial 22.73% Minimal 25.00% 50.00% -25.00% Substantial 37.50% Partial 50.00% 25.00% 25.00% Substantial 37.50% Partial 0.00% 50.00% -50.00% Partial 25.00% Minimal
Statutory
Approval
Related
Political 100.00% 90.90% 9.10% Substantial 95.45% Substantial 87.50% 100.00% -12.50% Substantial 93.75% Substantial 100.00% 100.00% 0.00% None 100.00% Total 100.00% 100.00% 0.00% Total 100.00% Total
Related
Construction 45.45% 45.45% 0.00% Total 45.45% Partial 62.50% 75.00% -12.50% Substantial 68.75% Substantial 75.00% 75.00% 0.00% None 75.00% Substantial 100.00% 100.00% 0.00% Total 100.00% Total
Related
Operational 100.00% 100.00% 0.00% Total 100.00% Total 100.00% 100.00% 0.00% Total 100.00% Total 100.00% 100.00% 0.00% Total 100.00% Total 0.00% 0.00% 0.00% Total 0.00% None
Related
Table 8.6 Comparison of missed risk events and differentiated risk perceptions.
teams should therefore incorporate specialist roles with relevant heuristics in contract pro-
curement and execution, similar to the particular project setting. And in order to achieve
this, there is the need for further empirical investigations to critically map out the heuris-
tics boundaries of the different specialist roles and also to develop a model for identifying
the heuristics indicators for different construction projects settings, to ensure compatibility
in the affective specialist-role stimulus of project team members and their specific project
setting. The intuitive investigative context of the empirical evidence, in the light of the
previous theoretical review on the operational variations between the different systems of
thinking and decision-making (Kahneman 2011 – discussed in Chapter 5), suggests that the
validity of the research empirical findings and potential benefits may be limited to intuitive
decision-making settings.
intuitive processing of quantity data formats (Akintoye and Macleod 1997; Lyons and Skit-
more 2004; Kululanga and Kuotcha 2010; Bowden et al. 2001; Lock 2003). This is in the light
of the prevailing behaviour science theory on the differences in structure and processes of
the intuitive and rational systems of thinking and decision-making (Kahneman 2011), and
the need for complementarity in system transformational processes (Walker 2015).
The following underlining theoretical issues were investigated:
Specialist role
Specialist role Case 1 Case 2 Case 3 Case 4 average score
120
100
80
60
40
20
0
Known- Known- Applied- Applied- Preferred- Preferred-
intuitive rational intuitive rational intuitive rational
Case 1 83.72 16.28 96.3 3.7 88.89 11.11
Case 2 74.51 25.49 91.43 8.57 92.59 7.41
Case 3 73.33 26.67 81.08 18.92 79.25 20.75
Case 4 85.71 14.29 87.5 12.5 93.65 6.35
Cross-case 79.3 20.7 89.1 10.9 88.6 11.4
Figure 8.1 Cross-case trend analysis of the known, applied, and preferred risk management
techniques.
20 years have been solely intuitive learning. The deviation of the two quantity surveyors
may be partly due to the mathematical orientation of the specialist – role risk management
practices (Edwards and Bowen 1998). The marginal inverse relationship between the study
participants’ years of industrial experience and their level of formal risk management train-
ing may have resulted from the contemporary changing perspectives in organisational risk
management strategies, with emphasis on formal risk management competencies (Loose-
more et al. 2006; Abderisak and Lindahl 2015).
The cross-case empirical findings on the risk management techniques have also revealed
high incidence of intuitive practices consistent with the current theoretical evidence
(Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha 2010).
The dominance of the intuitive decision processing system was very elaborate in the
responses from all the specialist roles and cases. The case 1 project engineer even suggested
that the intuitive approach is the only option available to the construction industry. As
many as 16 out of the 25 study participants (representing 64% – nine participants with
industrial experience equal or over 20 years; seven participants with industrial experience
under 20 years) confirmed sole reliance on intuitive risk management techniques. The
comparatively higher incidence of experienced study participants relying solely on intu-
itive techniques also appears consistent with the findings from a previous empirical study
which identified experienced non-graduate project managers relying solely on intuition
for their risk identifications (Maytorena et al. 2007).
The research findings have also discovered strategic differences in the risk management
practices of the private and public sectors cases. The private sector cases (case study
8.4 Research Proposition 3: Findings 183
projects 1 and 2) practised internal risk management using subsidiary contractor com-
panies, whilst the public sector cases (case study projects 3 and 4) on the other hand
applied external risk management through outsourcing to external contractors. The
empirical observation appears consistent with the wider industry practices of public
sector clients’ immense reliance on the design and build procurement route to transfer
project risk management responsibilities to third-party contractors, using contractual tools
(Latham 1994; Akintoye and Macleod 1997). An empirical investigation on the contractual
networks on four construction developments in 2012 also discovered evidence of internal
risk management on the private sector developments, using internal subsidiary companies
acting as principal contractors (Pryke 2012). The contracts manager from case 2 who was
also the project director similarly confirmed that most private developers do not follow
structured risk management systems. His observation over 36 years of industry experience
has been the use of internal subsidiary contractors to complete a risk register at project
commencement to satisfy corporate due diligence with subsequent reliance on personal
experiences and gut feelings to manage day to day project risk events. The strategic
differences between the private and public sector risk management approaches could be
described as two strands of intuitive construction risk management systems.
The trend analysis of the known, applied, and preferred risk management techniques has
been presented in Figure 8.1. The empirical evidence has revealed consistent high responses
for intuitive risk management techniques across the three assessment variables, which sug-
gests prolonged existence for the current intuitive risk management practices.
A significant discovery is the variance between the percentages of rational techniques
known, applied and preferred. The cross-case average scores have demonstrated that out of
the 20.7% ‘rational’ segment of the ‘known’ risk management techniques, only 52.66% are
being applied on the case study projects, representing 10.9% of the responses on ‘applied’
risk management techniques. There is however, a marginal increase of 55.1% preference
to apply the ‘known’ rational techniques on a future project, representing 11.4% of the
responses on ‘preferred’ risk management techniques. The marginal increase in preference
for future application of rational techniques also appears consistent with the previous the-
oretical discussions in Chapter 4, concerning the changing perspectives of organisational
risk management strategies with increasing emphasis on formal competencies to manage
project risk internally (Loosemore et al. 2006; Abderisak and Lindahl 2015). The empirical
evidence from case studies 1 and 3 exhibited the cross-case average trend. The findings from
case studies 2 and 4 on the other hand revealed marginal decreases in the preference for
future application of the ‘known’ rational techniques, compared to the percentages which
are currently being applied. The deviation in case study 2 may have been caused by the com-
bined influence of the low level of training in rational risk management techniques, and
the group’s extensive years of industrial experience observed in Table 7.19. The deviation
in case study 4 likewise may have been caused by the group’s extensive years of industrial
experiences (average of 23 years), which as previously discussed, tends to exhibit marginal
inverse correlation to rational decision-making systems.
Notwithstanding the above, the comparatively higher preference for future application
of the known ‘rational’ techniques over the current level of ‘rational’ application observed
in case studies 1 and 3, and the cross-case average, suggests the possibility of other extra-
neous causes for the high incidence of intuitive risk management practices, beside the
184 8 Application
90
80
70
60
50
40
30
20
10
0
Known- Known- Applied- Applied- Preferred- Preferred-
qualitative quantitative qualitative quantitative qualitative quantitative
Figure 8.2 Cross-case trend analysis of the known, applied, and preferred risk management data
presentation formats.
8.4 Research Proposition 3: Findings 185
preference for future application of quantitative data presentation formats also appears
consistent with the previous theoretical discussion in Chapter 4 concerning the changing
perspective of organisational risk management strategies, with increasing emphasis on
formal risk management competencies (Loosemore et al. 2006; Abderisak and Lindahl
2015).
The empirical evidence from case studies 2 and 4 exhibited the cross-case average trend.
The findings from case studies 1 and 3 on the other hand revealed marginal decrease in
the preference for future application of the ‘known’ rational techniques, compared to the
percentages which are currently being applied. The deviation in case studies 1 and 3 may
have been caused by the combined influence of the low level of training in rational risk
management techniques, and the groups’ extensive years of industry experience observed
in Table 8.7.
The empirical evidence of high reliance and preference for qualitative data presentation
formats across the four cases, at first glance may appear fully complimentary to the other
empirical evidence of intuitive dominated risk management techniques observed under
theoretical issue 6; as they both relate to the fast system of thinking and decision-making
(Kahneman 2011). Nevertheless, analysis of the variance between the proportions of
applied systems of risk management techniques presented under theoretical issue 6 (89.1%
intuitive versus 10.9% rational), against the applied systems of risk data formats (72.37%
qualitative versus 27.63% quantitative), suggests theoretically incompatible practices of
intuitive processing of approximately 60% of the applied quantitative data formats. This
was validated in the responses of the project manager from case 1; the contracts manager
from case 3; and the contracts manager and project manager from case 4, who after
confirming sole reliance on intuitive techniques under theoretical issue 6, also admitted
quantitative probability data application under theoretical issue 7. The empirical obser-
vation appears consistent with the existing theoretical evidence of intuitive processing of
quantitative statistical and probability risk data (Akintoye and Macleod 1997; Lyons and
Skitmore 2004; Kululanga and Kuotcha 2010; Bowden et al. 2001; Lock 2003). The research
theoretical review on the evolution of the UK construction industry’s risk management
systems from the initial risk conceptualisation within the insurance industry (Luhmann
2005; Smith et al. 2006 – discussed in Chapter 3) may hold some clue to how we got
into the current twist. Risk management practices within the insurance industry is very
structured and mathematically oriented (Everis 2009). The transition of risk management
systems from the UK insurance industry into the allied construction industry (Smith et al.
2006) therefore may have involved corresponding transmission of quantitative insurance
risk data presentation formats, whilst maintaining the traditional intuitive construction
decision processing approach.
The above empirical observations, coupled with the previous theoretical review on the
variance between the rational and intuitive decision-making systems (Slovic et al. 2010;
Kahneman 2011 – discussed in Chapter 5) and the need for complementarity between a
decision processing system and its data presentation format (Kahneman 2011; Walker 2015)
suggest the likelihood of systematic errors in the risk evaluations of some of the study par-
ticipants. The empirical evidence of low accuracy in intuitive processing of quantitative
statistical and probability data formats observed under theoretical issues 8 and 9 provides
the substantiation.
186 8 Application
Specialist role
Specialist role Case 1 (correct responses) Case 2 (correct responses) Case 3 (correct responses) Case 4 (correct responses) average score
Project manager 1 out of 4 (25%) 0 out of 4 (0%) 0 out of 4 (0%) 1 out of 4 (25%) 12.5%
Experience: 20 yr Experience: 16 yr Experience: 20 yr Experience: 21 yr 19.25
Training: 2 Training: 3 Training: 2, 3 Training: 2
Contracts manager 1 out of 4 (25%) 1 out of 4 (25%) 1 out of 4 (25%) 1 out of 4 (25%) 25%
Experience: 37 yr Experience: 36 yr Experience: 16 yr Experience: 10 yr 24.75
Training: 2 Training: 2 Training: 3 Training: 2, 3
Technical manager 1 out of 4 (25%) 1 out of 4 (25%) 0 out of 4 (0%) 16.67%
Experience: 20 yr Experience: 15 yr Experience: 19 yr 18
Training: 2, 3 Training: 2, 3 Training: 2, 3, 4
Architect 0 out of 4 (0%) 0 out of 4 (0%) 0 out of 4 (0%) 1 out of 4 (25%) 6.25%
Experience: 15 yr Experience: 6 yr Experience: 12 yr Experience: 18 yr 12.75
Training: 2, 3 Training: 2 Training: 2 Training: 2, 3
Engineer 2 out of 4 (50%) 2 out of 4 (50%) 0 out of 4 (0%) 0 out of 4 (0%) 25%
Experience: 14 yr Experience: 7 yr Experience: 19 yr Experience: 20 yr 15
Training: 2, 3 Training: 2, 3 Training: 2 Training: 2
Quantity surveyor 2 out of 4 (50%) 1 out of 4 (25%) 1 out of 4 (25%) 0 out of 4 (0%) 25%
Experience: 10 yr Experience: 36 yr Experience: 25 yr Experience: 30 yr 25.26
Training: 2, 3, 4 Training: 2, 3 Training: 3 Training: 2
Client 0 out of 4 (0%) 0 out of 4 (0%) 0%
Experience: 16 yr Experience: 46 yr 31
Training: 2 Training: 2
Case average score 29.17% 20.83% 8.33% 10.71% 17%
19 19 18 23 20
Specialist role
Specialist role Case 1 (correct responses) Case 2 (correct responses) Case 3 (correct responses) Case 4 (correct responses) average score
Project manager 0 out of 3 (0%) 0 out of 3 (0%) 2 out of 3 (66.7%) 0 out of 3 (0%) 16.67%
Experience: 20 yr Experience: 16 yr Experience: 20 yr Experience: 21 yr 19.25
Training: 2 Training: 3 Training: 2, 3 Training: 2
Contracts manager 0 out of 3 (0%) 0 out of 3 (0%) 3 out of 3 (100%) 2 out of 3 (66.7%) 41.67%
Experience: 37 yr Experience: 36 yr Experience: 16 yr Experience: 10 yr 24.75
Training: 2 Training: 2 Training: 3 Training: 2, 3
Technical manager 0 out of 3 (0%) 0 out of 3 (0%) 0 out of 3 (0%) 0%
Experience: 20 yr Experience: 15 yr Experience: 19 yr 18
Training: 2, 3 Training: 2, 3 Training: 2, 3, 4
Architect 0 out of 3 (0%) 1 out of 3 (33.3%) 0 out of 3 (0%) 2 out of 3 (66.7%) 25%
Experience: 15 yr Experience: 6 yr Experience: 12 yr Experience: 18 yr 12.75
Training: 2, 3 Training: 2 Training: 2 Training: 2, 3
Engineer 2 out of 3 (66.7%) 0 out of 3 (0%) 0 out of 3 (0%) 0 out of 3 (0%) 16.67%
Experience: 14 yr Experience: 7 yr Experience: 19 yr Experience: 20 yr 15
Training: 2, 3 Training: 2, 3 Training: 2 Training: 2
Quantity surveyor 2 out of 3 (66.7%) 0 out of 3 (0%) 2 out of 3 (66.7%) 3 out of 3 (100%) 58.33%
Experience: 10 yr Experience: 36 yr Experience: 25 yr Experience: 30 yr 25.26
Training: 2, 3, 4 Training: 2, 3 Training: 3 Training: 2
Client 0 out of 3 (0%) 0 out of 3 (0%) 0%
Experience: 16 yr Experience: 46 yr 31
Training: 2 Training: 2
Case average score 22.22% 5.56% 38.89% 28.57% 24%
19 19 18 23 20
The cross-case average scores have revealed just 24% accuracy in the total responses, with
the public sector cases (3 and 4) with their external risk management strategy performing
slightly better than the private sector cases (1 and 2) with their internal risk management
strategy, which appears contrary to the previous empirical findings presented in Table 8.8.
The inconsistences in market sector performance realised from the two assessment settings
suggest that the observed patterns of superior performances may have been influenced
by other extraneous factors unrelated to construction risk management strategy affective
heuristics.
The cross-case analysis of the average scores from the different specialist roles have also
revealed superior performance by the contracts managers and quantity surveyors, which
may be due to the comparatively mathematical orientation of the risk management prac-
tices of those specialist roles (Edwards and Bowen 1998) and the consequential impact on
their store of affective heuristics (Slovic et al. 2010). The empirical observation also appears
consistent with the present behaviour science evidence on the ability of experts to rely on
their professional experiences to achieve accurate intuitive judgement (Kahneman 2011).
Analysis of the cross-case empirical findings against the study participants’ years of
industrial experience and level of training in construction risk management techniques
also did not reveal any significant pattern, not even for those with some form of rational
risk management training and minimal industrial experience. In the light of the other
empirical evidence of intuitive dominated risk management practices observed under the-
oretical issue 6, coupled with the research theoretical review on the inherent psychological
difficulties in intuitive processing of quantitative data (Tversky and Kahneman 1982a,b,c;
Kahneman and Tversky 1982a,b; Kahneman 2011), the absence of significant relational
pattern between the study participants’ level of rational risk management training and
their level of accuracy in the risk assessment of probability data may have been influenced
by the applied intuitive decision-making system. This was also evident in the empirical
findings from the direct observations, where some of the study participants displayed signs
of emotional discomfort during the interrogations.
Case 1 (correct Case 2 (correct Case 3 (correct Case 4 (correct Specialist role average
Specialist role responses) responses) responses) responses) score
Project manager Quantitative: Wrong Quantitative: Wrong Quantitative: Correct Quantitative: Wrong Quantitative: 25%
Qualitative: Correct Qualitative: Wrong Qualitative: Wrong Qualitative: Wrong Qualitative: 25%
Experience: 20 yr Experience: 16 yr Experience: 20 yr Experience: 21 yr 19.25
Training: 2 Training: 3 Training: 2, 3 Training: 2
Contracts manager Quantitative: Wrong Quantitative: Wrong Quantitative: Wrong Quantitative: Wrong Quantitative: 0%
Qualitative: Correct Qualitative: Wrong Qualitative: Correct Qualitative: Correct Qualitative: 75%
Experience: 37 yr Experience: 36 yr Experience: 16 yr Experience: 10 yr 24.75
Training: 2 Training: 2 Training: 3 Training: 2, 3
Technical manager Quantitative: Correct Quantitative: Wrong Quantitative: Wrong Quantitative: 33.3%
Qualitative: Correct Qualitative: Correct Qualitative: Correct Qualitative: 66.7%
Experience: 20 yr Experience: 15 yr Experience: 19 yr 18
Training: 2, 3 Training: 2, 3 Training: 2, 3, 4
Architect Quantitative: Wrong Quantitative: Wrong Quantitative: Correct Quantitative: Correct Quantitative: 50%
Qualitative: Correct Qualitative: Correct Qualitative: Correct Qualitative: Correct Qualitative: 100%
Experience: 15 yr Experience: 6 yr Experience: 12 yr Experience: 18 yr 12.75
Training: 2, 3 Training: 2 Training: 2 Training: 2, 3
Engineer Quantitative: Wrong Quantitative: Wrong Quantitative: Wrong Quantitative: Wrong Quantitative: 0%
Qualitative: Correct Qualitative: Correct Qualitative: Correct Qualitative: Wrong Qualitative: 75%
Experience: 14 yr Experience: 7 yr Experience: 19 yr Experience: 20 yr 15
Training: 2, 3 Training: 2, 3 Training: 2 Training: 2
Quantity surveyor Quantitative: Wrong Quantitative: Wrong Quantitative: Wrong Quantitative: Correct Quantitative: 25%
Qualitative: Correct Qualitative: Correct Qualitative: Wrong Qualitative: Correct Qualitative: 75%
Experience: 10 yr Experience: 36 yr Experience: 25 yr Experience: 30 yr 25.26
Training: 2, 3, 4 Training: 2, 3 Training: 3 Training: 2
Client Quantitative: Wrong Quantitative: Wrong Quantitative: 0%
Qualitative: Correct Qualitative: Correct Qualitative: 100%
Experience: 16 yr Experience: 46 yr 31
Training: 2 Training: 2
Case average score Quantitative: 16.7% Quantitative: 0% Quantitative: 33.3% Quantitative: 28.6% Quantitative: 20%
Qualitative: 100% Qualitative: 66.7% Qualitative: 66.7% Qualitative: 71.4% Qualitative: 76%
19 19 18 23 20
the qualitative assessments across all the specialist roles. The evidence of lower accu-
racy in the intuitive processing of quantitative assessment appears consistent with the
other empirical evidence of low-level accuracy in statistics and probability assessment
observed under Theoretical Issue 8, and the previous theoretical review on the prevailing
behavioural science evidence discussed in Chapter 5 (Slovic et al. 2010; Kahneman 2011).
Another significant empirical discovery has been the composition of the wrong responses
selected under the quantitative assessments. 18 out of the 20 wrong respondents selected
the probability equation expressed in percentages, representing 90% of the wrong responses.
Further questioning on the rationale behind the preference for the ‘percentage equation’
format revealed the following responses:
1) Percentages are what we are familiar with from our primary and secondary educa-
tion – eight responses.
2) Percentages are what we use in our everyday working life – seven responses.
3) Percentages are easy to understand – three responses.
The above responses suggest that the participants who selected the percentage equation
were simply relying on their availability heuristics to intuitively generate the answers (Kah-
neman 2011). The observed behavioural pattern therefore supplements the other empirical
evidence discussed under theoretical issue 3.
Analysis of the cross-case empirical findings against the study participants’ years of
industrial experience and level of training in construction risk management techniques
again did not reveal any significant pattern, not even for those with some form of formal
risk management training and minimal industrial experience. Against the background of
the previous evidence of intuitive dominated risk management practices (theoretical issue
6) and the theoretical evidence on the psychological difficulties in intuitive processing
of quantitative data (Tversky and Kahneman 1982a,b,c; Kahneman and Tversky 1982a,b,
Kahneman 2011 – discussed in Chapter 5), the absence of significant relational pattern
between the study participants’ level of rational risk management training and their level
of accuracy in the quantitative assessments may likewise be attributed to the emotive
effect of the applied intuitive investigative approach. The direct observation of emotional
uneasiness during the interrogations provides the substation.
Specialist role
Case 1 (correct Case 2 (correct Case 3 (correct Case 4 (correct average score –
Specialist role responses) responses) responses) responses) insensitivity
Specialist role
Case 1 (correct Case 2 (correct Case 3 (correct Case 4 (correct average score –
Specialist role responses) responses) responses) responses) insensitivity
The empirical findings, though generally consistent across all the four cases, have been
more elaborate in the private sector cases, with their application of internal risk manage-
ment strategy. The level of insensitivity among the study participants from the private sector
cases (1 and 2) is 83.33%, against the public sector cases (3 and 4) insensitivity scores of
50% and 57.14%, respectively. The cross-case analysis of the average scores from the differ-
ent specialist roles have also revealed an inverse correlation with the pattern of empirical
evidence on the level of understanding of probability discussed under theoretical issue 8.
The contracts manager and quantity surveying specialist roles which exhibited compara-
tive superior understanding of probability data under theoretical issue 8, have also exhib-
ited comparative lower insensitivity to the varied hypothetical probability predictions of
asbestos contamination. On the other hand, the remaining specialist roles which demon-
strated lower understanding of the probability principle under theoretical issue 8 have like-
wise demonstrated higher insensitivity to the diverse hypothetical asbestos contamination
predictions. The inference being that the extent of insensitivity may inversely correlate to
the level of understanding of the probability concept.
Analysis of the cross-case empirical findings against the study participants’ years of
industrial experience and level of training in construction risk management techniques
once again did not reveal any significant pattern, not even for those with some form of
rational risk management training and minimal industrial experience. In the light of the
previous empirical evidence of intuitive dominated risk management practices, coupled
with the theoretical discussions on the inherent psychological difficulties in intuitive
processing of quantitative data (Tversky and Kahneman 1982a,b,c; Kahneman and Tversky
1982a,b, Kahneman 2011), the absence of significant relational pattern between the study
participants’ level of rational risk management training, and their level of insensitivity
to the hypothetical probability risk estimates, may also be attributed to the emotional
effects from the intuitive decision-making system. This was also evident in the empirical
findings from the direct observations where some of the study participants became very
poignant during the interrogations. The project manager from case 1 and the client
development manager from case 3 responded that they will take no chances with the lives
of their workforce irrespective of the level of probability of asbestos risk. The contracts
manager from case 2 in particular descended into intense emotions after listening to the
theoretical settings of the interview question and responded that his risk response to
asbestos contamination will always be avoidance, irrespective of the probability estimate.
Upon probing, he explained that he had lost his father a year ago from mesothelioma. His
zero tolerance to asbestos risk most likely was precipitated by an inducement of a strong
affective feeling associated with his previous bitter experience of observing his father dying
from asbestos cancer (Bateman et al. 2010; Slovic et al. 2010).
In summary, the empirical evidence from theoretical issues 6–10 have supported research
proposition 3. The observation of systematic errors arising from the theoretical imbalances
in the construction risk management decision-making systems, coupled with the direct
observations of emotional discomfort from some of the study participants during the statis-
tics and probability risk assessments, has collectively confirmed the complications in intu-
itive processing of quantitative data. The realisation of consistent empirical support across
the different specialist roles and different assessment settings applied on the four cases have
also confirmed theoretical replication for the study’s theoretical predictions, and subse-
quently analytical generalisation for research proposition 3.
196 8 Application
8.5 Summary
Chapter 8 has examined the extent to which the empirical evidence from the cross-case
analysis reflected the three research theoretical propositions. The findings are presented
below:
system. In a situation where the different specialist roles decide to adopt different risk
management systems, there could be cross-purpose strategies which could be damaging to
project success (Pryke and Smyth 2006; Cagno et al. 2007; Arthur and Pryke 2013). Even in
a situation where a joint risk management system is practised, the failure to recognise and
integrate the differences in affective heuristics of the assembled specialist roles could lead
to conflict (Mullins 2005). The validation of the research novel theoretical philosophy for
risk perception categorisation therefore suggests the need for expansion of the integrative
role of construction project management (Walker 2007, 2015) to encompass both the
technical and affective heuristics variances of the different specialist roles to foster efficient
risk management systems, and ultimately, project success (Pryke and Smyth 2006; Cagno
et al. 2007).
professional expertise and experience of the assembled project team within the construc-
tion delivery system. Where the project team lacks resources in a specific specialist role
affective heuristic, the absence of an associated availability and representative heuristics
(Kahneman 2011) may potentially constrain the identification and management of related
risk events (Benthin et al. 1993). Even within a multi-disciplined project team, the exclusion
of a specialist role from the risk management process may likewise weaken the robustness
of the risk identification and treatment processes, especially for the risk events relating to
the excluded specialist role affective heuristics. The empirical findings therefore suggest the
need for careful selection of project team members with the appropriate grounded heuris-
tics, to ensure effective risk management systems. The formation of project teams should
therefore incorporate specialist roles with relevant heuristics in contract procurement and
execution, appropriate for comparable project settings. And in order to achieve this, there is
the need for further empirical investigations to critically map out the heuristics boundaries
of the different specialist roles, and also to develop a model for identifying the heuristics
indicators for different construction project settings, to ensure compatibility in the affective
specialist-role stimulus of project team members and their specific project setting (Arthur
2018).
(Edwards and Bowen 1998), and the resultant impact on their store of affective heuris-
tics (Slovic et al. 2010). The observation from the contracts manager, and quantity sur-
veyor specialist roles also appears consistent with the present empirical evidence from the
behavioural sciences regarding the ability of experts to rely on their professional experiences
to achieve accurate intuitive judgment (Kahneman 2011).
The empirical evidence from theoretical issue 9 has likewise demonstrated superior
level of accuracy for qualitative assessments compared to quantitative assessments, which
appears consistent with the prevailing behaviour science theories (Slovic et al. 2010;
Kahneman 2011).
Lastly, the empirical evidence from theoretical issue 10 has revealed significant insensi-
tivity for probability predictions of emotive events, in intuitive evaluations. This also seems
consistent with the prevailing behaviour sciences theories (Slovic et al. 2010). Analysis of
the study participants’ responses has further discovered a negative correlation between the
extent of insensitivity and the level of understanding of the probability concept.
The empirical findings from theoretical issues 6–10 have collectively provided substan-
tial support for the research proposition 3. The observation of systematic errors arising from
the theoretical imbalances in the construction risk management decision-making systems,
coupled with the direct observations of emotional discomfort from some of the study par-
ticipants during the statistical and probability risk assessments, has collectively confirmed
the complications in intuitive processing of quantitative data. The realisation of consis-
tent empirical support across the different specialist roles and different assessment settings
applied on the four cases have also confirmed theoretical replication for the study’s theoret-
ical predictions, and subsequently analytical generalisation for research proposition 3.
The absence of significant differences between the findings from the private sector cases
(1 and 2) and public sector cases (3 and 4) with their different risk management strate-
gies, have confirmed independence of the empirical evidence outside the influence of risk
management strategy affective heuristics.
The realisation of consistent evidence across the different specialist roles and cases
has confirmed theoretical replication for the 10 theoretical issues (Rowley 2002), and
subsequently, analytical generation for the three research propositions (Eisenhardt 1989;
Yin 2014). It must however be emphasised that the intuitive investigative context of
the empirical evidence, in the light of the previous analytical review on the operational
variations between the different systems of thinking and decision-making (Kahneman
2011 – discussed in Chapter 5) suggests the validity of the research findings and potential
benefits, including the novel theory for risk perception categorisation, may be limited to
applications under intuitive decision-making.
The research findings have in effect challenged the prevailing construction risk manage-
ment practice of intuitive processing of statistical and probability data presentation formats
and further established a logical foundation for the rethinking of construction risk man-
agement practices to ensure complementarity between the decision processing approach
and data presentation formats. The previous analytical review on the differences in opera-
tion of the rational and intuitive decision-making systems (Kahneman 2011 – discussed in
Chapter 5) coupled with the theoretical evidence on the need for complementarity between
system components to foster the achievement of system objectives (Cole 2004; Cole and
200 8 Application
Kelly 2020; Walker 2007, 2015 – discussed in Chapter 2) suggests the reasonable method-
ological remediation to be either revision in the construction risk management decision
processing approach to reflect proportional application of rational techniques to comple-
ment the quantitative properties of statistics and probability or alternatively, revision of the
data presentation formats to qualitative to align with the existing intuitive decision pro-
cessing practices. The other research empirical evidence of low competency in rational risk
management techniques observed under theoretical issue 6 and the high preference for
future application of subjective techniques and qualitative data observed under theoreti-
cal issues 6 and 7 suggests prolonged existence for the present intuitive construction risk
management decision processing approach. Revision of the data presentation formats to
qualitative therefore seems the most appropriate option. The theoretical evidence of spon-
taneous and effortless operation of the intuitive decision-making system (Kahneman 2011)
also suggests that adopting the above remedial approach will assist construction profes-
sionals to address the increasing client and regulatory demand for quick and accurate risk
responses using limited available data (Loosemore et al. 2006; Perlow et al. 2002). There
is however the need for further empirical investigates to test the appropriate qualitative
data presentation formats which will complement the present intuitive construction risk
management practices.
201
Conclusions
9.1 Summary
The prevailing theoretical evidence of intuitive dominated construction risk management
practices (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha
2010) suggests corresponding reliance on emotional stimulus in the decision-making
processes (Slovic et al. 2010). The current construction management research publications
however appear limited in exploring the psychology behind the intuitive risk identification
and treatment processes. The purpose of this book therefore has been to address the
resulting data gap, by drawing upon theories from systems thinking and behavioural
sciences to examine the behavioural patterns of intuitive construction risk management
decision-making systems in the context of the sources and structure of the risk perception
categorisations and the psychological issues involved in mixing tools and techniques from
different systems of thinking and decision-making.
The Part 1 has evaluated key concepts on intuitive risk management systems. The
theoretical framework centred on the social construction of risk within the construction
delivery process and the subjectivity of risk perceptions. The discussions have employed
systems thinking and behaviour science theories in reviewing key concepts includ-
ing general system principles and processes (Checkland 1999; Bertalanffy 1968, 2015;
Cole 2004; Carmichael 2006; Walker 2007, 2015; and Cole and Kelly 2020 – discussed
in Chapter 2); the generation of construction project risk events from systems differ-
entiation processes (Blanchard and Fabrycky 1998; Walker 2007, 2015; Slovic et al.
2010 – discussed in Chapter 2); risk etymology (Luhmann 2005; Japp and Kusche 2008;
The Risk Net 2020 – discussed in Chapter 3); the different risk conceptual interpretations
(Zinn 2008 – discussed in Chapter 3); the construction risk management process (Smith
et al. 2006; Dallas 2006; Walker 2007, 2015 – discussed in Chapter 4); risk perception forma-
tion from affect heuristics (Benthin et al. 1993; Damasio 2006; Slovic et al. 2010 – discussed
in Chapter 5); risk data presentation formats (Bowers 1994; Bowden et al. 2001; Slovic
et al. 2010 – discussed in Chapter 5); subjectivity of risk perceptions, and the inherent
biases of intuitive decision-making (Tversky and Kahneman 1982a,b,c; Kahneman and
Tversky 1982a,b; Kahneman 2011 – discussed in Chapter 5); and the generating processes
of differentiated risk perceptions (Carmichael 2006; Walker 2007, 2015; Slovic et al. 2010;
Kahneman 2011 – discussed in Chapter 5). The systematic evaluation of the above
theoretical concepts has strengthened their analytical interpretations.
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
202 9 Conclusions
The Part 2 has presented findings from a four-case empirical investigation evaluating
the theoretical and practical issues of the current intuitive construction risk management
practices. The research methodology focused on testing theoretical replication of the propo-
sitions developed in the Part 1 including the novel differentiated risk perception model.
A bespoke matching pattern analytical tool called the differentiated matrix has also been
developed. This tool was applied in the analysis of the research’s novel theoretical expla-
nation for risk perception categorisation, under proposition 1, and again, in demonstrating
the effects of excluding project subgroups from intuitive risk management processes, under
proposition 2. The inherent dynamics of the analytical tool, which enabled multiple appli-
cations within the different categorisation contexts suggest there could be potential for
extending it into other behavioural categorisation studies within construction management
research and other allied social science disciplines.
The specific empirical contributions from the three research propositions are as follows:
(Carmichael 2006; Walker 2007, 2015 – discussed in Chapter 2), it would be worth con-
ducting further empirical investigations to test their influence in risk perception categori-
sation at the construction project organisational level.
increasing client and regulatory demand to provide quick and accurate risk responses
using limited available data (Loosemore et al. 2006; Perlow et al. 2002). There is however
the need for further empirical investigations into the appropriate qualitative data pre-
sentation formats which will be compatible with the existing intuitive construction risk
management practices. As a start, it may be worth examining the reliability of fuzzy sys-
tem linguistic data application (Byrne 1995), within the present intuitive construction
risk management decision-processing systems.
risk management researchers, training providers, and professionals support the paradigm
transition by adopting the novel philosophy for risk perception categorisation based
on the differences in affective heuristics of the differentiated specialist roles involved in
construction project delivery; restructuring the existing construction risk management
training programmes to emphasise practises which promote compatibility between the
decision processing vehicle and the data presentation format; and also join the exploratory
wagon for the discovery of appropriate qualitative data presentation formats to align with
the current intuitive decision processing practices.
207
Appendix A
Research strategy
Theoretical Related How does Population for and data
issues questions it work data collection collection approach
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
208 Appendix A Research Design – Theory, Methodology, and Field Questions
Research strategy
Theoretical Related How does Population for and data
issues questions it work data collection collection approach
Research strategy
Theoretical Related How does Population for and data
issues questions it work data collection collection approach
3. Perceptions are Examine the Theoretical issue to Project team Research strategy:
formed from the influence of be confirmed if the members involved in Case study
representative and grounded heuristics participants’ risk design development
availability in the risk identification on and risk management Data collection
heuristics identification the hypothetical decision-making, approach: Direct
(Kahneman 2011; process using a project scenarios including project observations and
Slovic et al. 2010) hypothetical correspond to their managers, contracts semi-structured
construction project availability managers, quantity interviews to
scenarios (i.e. if the heuristics (evident surveyors, technical examine correlation
participant’s risk from their coordinators, between
identification has responses on the architects, engineers participants risk
been done common and clients drawn perception and
intuitively) construction project from four case study their grounded
(Refer to IQ 22–24) risk events) projects (two public heuristics evident
sector and two in response to
private sector) questions 22 and 23
Appendix A Research Design – Theory, Methodology, and Field Questions 209
Research strategy
Theoretical Related How does Population for and data
issues questions it work data collection collection approach
4. The inherent Testing subjectivity Theoretical issue to Project team Research strategy:
subjectivity of and manipulative be confirmed if members involved in Case study
perceptions make it influence from exposure to design development
manipulative and perceptions additional and risk management Data collection
less rational. (influence of information decision-making, approach: Direct
Additional exposed heuristics (a schedule of including project observations and
information could in manipulating construction risk managers, contracts semi-structured
change risk risk perception/ sources) changes managers, quantity interviews to
perception even analysis) the participant’s surveyors, technical examine if exposure
where the base (Refer to IQ 24 risk perception/ coordinators, to additional
situation has not and 25) analysis architects, engineers information can
changed – best case and clients drawn influence/
scenarios which from four case study manipulate risk
lead to gaps between projects (two public perception/analysis
pre-commencement sector and two
estimates and actual private sector)
figures at
completion stage
(Flyvbjerg et al.
2003; Alhakami and
Slovic 1994; Slovic
and Peters 2006)
5. An uncertainty Analysis of Theoretical issue to Project team Research strategy:
becomes a risk event responses from be confirmed if members involved in Case study
to a subject if its theoretical issues 2a participants fail to design development
occurrence can and 2b to verify if identify risk events and risk management Data collection
potentially affect the the absence of outstand their spe- decision-making, approach: Direct
objective(s) of the relevant grounded cialist/professional including project observations and
subject (Loosemore heuristics can affect areas (associated managers, contracts semi-structured
et al. 2006), risk identification/ with lack of related managers, quantity interviews to
suggesting that analysis (Refer to grounded heuristics) surveyors, technical examine if the
where there is no IQ 24) coordinators, missed risk events
possibility of architects, engineers can be attributed to
impact, the subject and clients drawn the lack of relevant
may fail to recognise from four case study grounded heuristics
the uncertainty as a projects (two public
risk event sector and two
private sector)
Research proposition 3: Statistical and probability risk data presentation and analysis are
less likely to be understood and applied in intuitive risk management systems compared to
the use of qualitative data presentation formats and subjective assessment techniques.
210 Appendix A Research Design – Theory, Methodology, and Field Questions
Research strategy
Theoretical Related How does Population for and data
issues questions it work data collection collection approach
6. Research findings Assess the level of Theoretical issue to Project team Research strategy:
have identified knowledge and be confirmed if the members involved in Case study
construction risk application of risk empirical evidence design development
management to be management reveals preference and risk management Data collection
dominated by techniques for intuitive risk decision-making, approach: Direct
people without (Refer to IQ 7–11) management including project observations and
formal training in approaches over managers, contracts semi-structured
risk management rational tools and managers, quantity interviews to assess
systems using techniques surveyors, technical competency in
intuition and coordinators, probability tools
heuristics (Akintoye architects, engineers
and Macleod 1997; and clients drawn
Lyons and Skitmore from four case study
2004; Kululanga and projects (two public
Kuotcha 2010) sector and two
private sector)
7. Psychometric Assess the level of Theoretical issue to Project team Research strategy:
research findings knowledge and be confirmed if the members involved in Case study
suggest that most application of empirical evidence design development
people relate better statistical models reveals preference and risk management Data collection
to narratives, and tools for qualitative data decision-making, approach: Direct
models, illustrations (Refer to IQ 12–14) presentation including project observations and
and anecdotal formats over managers, contracts semi-structured
simulations better statistical data and managers, quantity interviews to assess
than statistical data quantitative models surveyors, technical competency in
(Slovic et al. 2010) coordinators, probability tools
architects, engineers
and clients drawn
from four case study
projects (two public
sector and two
private sector)
8. Most people do Assess the level of Theoretical issue to Project team Research strategy:
not trust, knowledge and be confirmed if the members involved in Case study
understand or use application of empirical evidence design development
statistics and statistical and reveals low level of and risk management Data collection
probability probability tools knowledge and decision-making, approach: Direct
estimates in their (Refer to IQ 15–18) application of including project observations and
risk management statistics and managers, contracts semi-structured
decision-making probability managers, quantity interviews to assess
(March and Shapira assessments surveyors, technical competency in
1987; Akintoye and coordinators, statistical models
Macleod 1997; architects, engineers and tools
Edwards and Bowen and clients drawn
1998; Lyons and from four case study
Skitmore 2004) projects (two public
sector and two
private sector)
Appendix A Research Design – Theory, Methodology, and Field Questions 211
Research strategy
Theoretical Related How does Population for and data
issues questions it work data collection collection approach
9. Most people relate Intuitive judgement Theoretical issue to Project team Research strategy:
better to qualitative of hypothetical risk be confirmed if the members involved in Case study
assessments than events expressed in empirical evidence design development
quantitative quantitative and reveals high and risk management Data collection
assessments (Slovic qualitative formats accuracy for risk decision-making, approach: Direct
et al. 2010) (Refer to IQ 19 analysis expressed including project observations and
and 20) in whole numbers, managers, contracts semi-structured
qualitative managers, quantity interviews to
statements, and surveyors, technical examine if different
scenarios, coordinators, probability
compared to risk architects, engineers predictions of the
analysis expressed and clients drawn likelihood of
in percentages, from four case study asbestos
decimals and ratios projects (two public contamination and
sector and two impact can
private sector) influence risk
response
10. Where a Assess the impact Theoretical issue to Project team Research strategy:
potential outcome of of strong affective be confirmed if the members involved in Case study
a future event stimuli on empirical evidence design development
evokes a strong probability reveals insensitivity and risk management
affective memory, evaluations to different ranges decision-making,
Data collection
the assessor may be (Refer to IQ of hypothetical including project
strategy: Direct
insensitive to the 21i–21iii) probability managers, contracts
observations and
probability predictions of managers, quantity
semi-structured
predictions on the asbestos surveyors, technical
interviews to assess
event (Slovic et al. contamination risk coordinators,
the comparative
2010) architects, engineers
effectiveness of
and clients drawn
instinctive risk
from four case study
analysis utilising
projects (two public
qualitative model
sector and two
over quantitative
private sector)
models
Interview Questions
Section A. General Details: Questions 1–5 Collects General Details of the Respondent
1 Please select the professional/specialist area that best describe your role in the con-
struction industry?
Architect (………)
Technical Co-ordinator/Manager (………)
Quantity Surveyor (………)
Structural Engineer (………)
Electrical Engineer (………)
212 Appendix A Research Design – Theory, Methodology, and Field Questions
3 Please select from below the type of organisation that best describe your area of
employment?
Client/Developer (………)
Contractor (………)
Consultant (………)
Others (please specify)…………………………………………………………………
6 How often do you get involved in the identification and management of project con-
straints and opportunities (as a project team member or participation in project risk
management workshop/brainstorming)?
Always (………)
Sometimes (………)
Sparsely (………)
Never (………)
Appendix A Research Design – Theory, Methodology, and Field Questions 213
7 Please indicate your level of education/training in project risk management tools and
practices?
a. No formal education or training in project risk management (………)
b. On the job informal experience (………)
c. Continuing professional developments (CPDs) and seminars on project risk man-
agement (………)
d. Short duration formal course (less than six months) (………)
e. University diploma/degree in Project Risk Management – e.g. BSc Construction Risk
Management, MSc Risk Management ( )
Please specify title of degree/diploma ……………………………………………… . . . ..
8 Which of the following risk management techniques have you heard of? (Please
indicate with a tick)
Risk register ………… . . . .. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) ………… Risk breakdown analysis ………… . . . ..
Worse/best case scenario ……………… Monte Carlo simulation/analysis ………… . . . .
Brainstorming………… . . . . Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ……… . . . .. Other (please specify) …………………………
9 Which of the following risk management techniques have you applied on this project?
(Please indicate with a tick)
Risk register ………… . . . .. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) ………… Risk breakdown analysis ………… . . . ..
Worse/best case scenario ……………… Monte Carlo simulation/analysis ………… . . . .
Brainstorming……………… Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ……… . . . . Other (please specify) …………………………
10 On a scale of 1–3 please rank the top three (3) techniques which you would
prefer to use on your future risk management practices? You may select from the
following list – Risk register, Probability matrix, Fault tree, Personal feelings and values,
Personal experiences (heuristics), Risk breakdown analysis, Worse/best case scenario,
Monte Carlo simulation/analysis, Brainstorming, Failure mode and effect analysis, Fish
bone, Cultural believes, and Gut feeling (please note: A scale of 1 represents the highest
preference, and a scale of 3 represents the lowest preference)
214 Appendix A Research Design – Theory, Methodology, and Field Questions
11 What are the factors influencing your preference of the above risk management
technique?
1st ranked technique ………………………………………………………………………
2nd ranked technique …………………………………………………………………….
3rd ranked technique ………………………………………………………………………
12 Which of the following forms of risk data presentations do you know? (Please tick all
the applicable)
Narrative …… . . . .. Probabilities ……………… . . . .
Qualitative statements ………… Whole numbers ………… . . . .
Graphs ……… . . . .. Ratios ……………………………
Fractions ……… . . . . Scenarios …… . . . .
Simulations ……… . . . .. Other (please specify) ……………
13 Which of the following forms of risk data presentations have you applied on this
project? (Please tick all the applicable)
14 On a scale of 1–3, please rank the top three (3) risk data presentation formats which
you would prefer to use on your future risk management practices? You may select
from the following list – Narratives, Probabilities, Qualitative statements, Whole num-
bers, Graphs, Ratios, Fractions, Scenarios, and Simulations (please note: A scale of 1
represents the highest preference, and a scale of 3 represents the lowest preference)
1st ranked …………………………………………………………………………………
2nd ranked …………………………………………………………………………………
3rd ranked …………………………………………………………………………………
Appendix A Research Design – Theory, Methodology, and Field Questions 215
15 Which of the following statistical symbols are you familiar with? Please indicate by
selecting from the following list: summation, probability density function, probability
function, sample mean, population mean, median, variance, standard deviation, sample
standard deviation, and don’t know
Symbol Name
P(A)
s
∑
Please use the hypothetical risk evaluations represented in the table below to answer
questions 16–18
A 1/7 30/35
B 97.9% 84%
C 1 : 9743 4:5
18 Which of the above risk events is certain to have the highest impact?
Risk event A ……………………………………………
Risk event B ……………………………………………
Risk event C ……………………………………………
All (risk events A, B, and C) …………………………
None of the risk events ………………………………
Don’t know ……………………………………………
19 Which of the following hypothetical project settings offer the highest probability
estimate (Indicate with a tick)
A. There is a 75.2% probability of half of the site being contaminated (………)
B. There is a 1 : 321 probability of the site being contaminated (………)
C. There is a 0.381 probability of the site being contaminated (………)
20 Which of the following hypothetical project settings offer the highest probability esti-
mate? (Indicate with a tick)
A. The probability of projects over running their initial budget is 3 out of 10 (………)
B. out of every projects overrun their initial budget
(………)
C. After reviewing the financial records of the four completed construction projects,
only project A recorded a final cost higher than the initial budget estimate (………)
Asbestos related illness has been a major cause of construction occupational death. Accord-
ing to the Health and Safety Executive in the United Kingdom, there has been a recent surge
in asbestos related illness and deaths (see the table and pictures below).
Appendix A Research Design – Theory, Methodology, and Field Questions 217
1968 153
2009 2321 411
2010 2347 412
2011 2291 429
2012 2535 464
Source: Shutterstock.com
218 Appendix A Research Design – Theory, Methodology, and Field Questions
21i Please kindly review the information above and complete the ‘Appropriate risk
response’ column of the ‘Hypothetical project risk evaluation matrix’ below. Please
select your response from any of the following options:
i. Appoint a licenced contractor to carry out further surveys to identify and remove
the asbestos before the commencement of construction works
ii. Carry on construction works at the non-contaminated area whilst carrying out
further investigations at the remaining areas to identify and remove any asbestos
residue
iii. Supply protective masks to all operatives and carry on with the construction works
iv. Carry on with the construction works and forget about any asbestos
21ii What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
(Please select your response from any of the options given under 21i.)
21iii What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
(Please select your response from any of the options given under 21i.)
Section G-1. Construction Project Risk Events – Part 1: Questions 22–24 Requires
Ranking of the Typical Construction Risk Events at the Different Phases of Project
Execution
22 Drawing from your professional experiences, what will you say are the factors likely
to affect project success (risk events) at the pre-construction phase (before the start of
actual construction on site)?
1. ..……… . . . .……………………………………………………………………………
2. …………………………………………………………………………………………
3. …………………………………………………………………………………………
23 Drawing from your professional experiences, what will you say are the factors likely to
affect project success (risk events) at the construction phase (during actual construc-
tion work on site)?
1. ..……… . . . .……………………………………………………………………………
2. …………………………………………………………………………………………
3. …………………………………………………………………………………………
24 The pictures below depict hypothetical sites (edged in bold front) being considered for
residential development. The project is at the inception stage with the following key
project objectives;
– Design: 50 units to meet minimum building regulation requirements, lifetime
homes, secured by design, and minimum 40% carbon reduction
220 Appendix A Research Design – Theory, Methodology, and Field Questions
Hypothetical site A: Brown field development within high density urban area
Hypothetical site B: Green field development within low density rural area
Please list the factors that are likely to affect the success of the projects (risk events)
during the pre-construction and actual construction phases.
1 1
2 2
3 3
4 4
5 5
Section G-2. Construction Project Risk Events – Part 2: Question 25 Examines the
Impact of Additional Information on Intuitive Risk Evaluation
Please refer to the table of construction risk sources below and answer question 25 (please note
the question is a repetition of question 24. You are kindly requested not to use the table of
construction risk sources to amend your previous responses to question 24. It is acceptable to
have different responses for the two identical questions). Please kindly select your responses
from the ‘Sources of risk’ column.
Construction project risk sources
25 The pictures below depict hypothetical sites (edged in red) being considered for
residential development. The project is at the inception stage with the following key
project objectives;
– Design: 50 units to meet minimum building regulation requirements, lifetime
homes, secured by design, minimum 40% carbon reduction
– Budget: Construction cost not to exceed £15 million
– Programme: To be completed within two years
Hypothetical site A: Brown field development within high density urban area
Please list the factors that are likely to affect the success of the projects (risk events)
during the pre-construction and actual construction phases.
1 1
2 2
3 3
4 4
5 5
26 We have now completed the interview questions. I would however like to give you
the opportunity if you have any comment to add. Your responses will be analysed and
included in the empirical investigation report. You may provide your contact details
if you wish to receive a copy of the executive summary of the report. Thank you for
taking the time to participate.
Comments: ………………………………………………………………………………
……………………………………………………………………………………………
Name: ……………………………………………………………………………………
Email: ……………………………………………………………………………………
Telephone number:………………………………………………………………………
225
Appendix B
Appendix B presents the empirical evidence from case study 2. The findings have been pre-
sented under three subsections corresponding to the three research propositions developed
in Section 6.2. The findings for the propositions have been further organised under separate
headings relating to the underlining theoretical issues. The presentation has adopted the
question and answer approach. The interview questions relating to each theoretical issue
have been described in Appendix A.
Theoretical Issue 1
– Theoretical issue 1: Systems differentiation also generate differences in risk perceptions
among the system components – different specialist roles (Walker 2007, 2015; Kahneman
2011; Slovic et al. 2010).
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
226 Appendix B Case 2 Data Presentation
The designers comprising the engineer and architect also generated risk perceptions in
the following order of ranking: design and quality related (first), financial and commercial
related (second), and construction related (third). The quantity surveyor’s generated
risk perceptions on the other hand were in the following order of ranking: financial and
commercial related (first), force majeure related (second), and design and quality related
(third).
Theoretical Issue 2
– Theoretical issue 2: The differences in objectives of differentiated subsystems
(Carmichael 2006; Loosemore et al. 2006) suggest that the impact of environmen-
tal forces will result in differences in affective heuristics, and therefore differences in
perceptions within the system components – different specialist roles.
Theoretical issue to be confirmed if the empirical evidence reveals similarities in the
respondents’ risk perceptions on hypothetical project settings and their specialist back-
ground.
The related interview question is 24 (see Appendix A). The rationale behind this question
was to solicit the participants’ risk perceptions on two hypothetical construction projects.
Interview Question 24 The pictures below depict hypothetical sites (edged in bold front)
being considered for residential development. The project is at the inception stage with the
following key project objectives:
– Design: 50 units to meet minimum building regulation requirements, lifetime homes,
secured by design, and minimum 40% carbon reduction.
– Budget: Construction cost not to exceed £15 million.
– Programme: To be completed within two years.
Hypothetical site A: Brown field development within high density urban area
Hypothetical site B: Green field development within low density rural area
Findings: The responses outline the participants’ risk perceptions on the hypothetical
projects. The risk events have been categorised using the table of construction project risk
events (see Table 7.3). The findings have been summarised in Tables B.3–B.5.
There is further comparison of the data from the two hypothetical projects to check for
consistency and also examine the pattern of the generated risk perceptions for evidence of
differentiation along the lines of the research participants’ respective specialist background.
The consistency analysis was also meant for filtering out the other extraneous risk forma-
tion variables which are unrelated to specialist role affective heuristics.
The empirical findings have revealed evidence of differentiation within the generated
risk perceptions from the two hypothetical projects, with the pattern of categorisation being
consistent with the specialist background of the study participants. The consistency analysis
has also revealed the following:
– Design and quality related risk events were perceived by the architect and engineer.
– Logistics and communication related risk events were perceived by the quantity surveyor
and technical manager.
– Financial and commercial related risk events were perceived by the quantity surveyor,
project manager, and engineer.
– Legal and statutory approval related risk events were perceived by the quantity surveyor,
project manager, and architect.
– Construction related risk events were also perceived by the project manager and contracts
manager.
B.1 Research Proposition 1: Findings 231
Case study 2
Cumulative Cumulative
for for
Risk perception PM-2 TM-2 CM-2 builders A-2 E-2 designers QS-2
Force majeure related
Political related
(Continued)
232 Appendix B Case 2 Data Presentation
Case study 2
Cumulative Cumulative
for for
Risk perception PM-2 TM-2 CM-2 builders A-2 E-2 designers QS-2
Construction related
Programming of construction Yes Yes Yes
works to achieve completion
date
Safety on site Yes Yes
Operational related
Evaluation of the evidence from the consistency analysis from the perspective of the dif-
ferent specialist roles has also revealed the following categorisations:
– The builders comprising the project manager, the technical manager, and the contracts
manager have exhibited diverse risk perceptions associated with logistics and communi-
cation, financial and commercial, legal and statutory, and construction related events.
– The designers comprising the architect and the engineer also displayed risk perceptions
relating to ground condition, design and quality, financial and commercial, and legal
and statutory approval events. Design and quality risk events were consistent in all the
responses.
– The quantity surveyor on the other hand has exhibited risk perceptions relating to finan-
cial and commercial, logistics and communication, and legal and statutory approval
events.
In summary, the empirical evidence from Theoretical Issues 1 and 2 have revealed sub-
stantial support for research proposition one. The findings on differentiation within the
study participants’ risk perceptions on the typical project risk events at the different con-
struction delivery phases and also under different hypothetical project settings have been
very consistent with the study’s theoretical predictions. The realisation of patterns of risk
perceptions correlating to the specialist background of the study participants has subse-
quently provided empirical support for the research novel proposition of systems differen-
tiation influencing risk perception categorisation within the construction project delivery
system.
Case study 2
Cumulative Cumulative
for for
Risk perception PM-2 TM-2 CM-2 builders A-2 E-2 designers QS-2
Force majeure related
Political related
(Continued)
234 Appendix B Case 2 Data Presentation
Case study 2
Cumulative Cumulative
for for
Risk perception PM-2 TM-2 CM-2 builders A-2 E-2 designers QS-2
Construction related
Programming construction Yes Yes
works to achieve project
completion date
Safety on site Yes Yes
Operational related
– 3. Perceptions are formed from the representative and availability heuristics (Kahneman
2011; Slovic et al. 2010).
– 4. The inherent subjectivity of perceptions makes it manipulative and less rational. Addi-
tional information could change risk perception even where the base situation has not
changed – best-case scenarios which lead to gaps between pre-commencement estimates
and actual values at completion stage (Flyvbjerg et al. 2003; Alhakami and Slovic 1994;
Slovic and Peters 2006).
– 5. An uncertainty becomes a risk event to a subject if its occurrence can potentially affect
the objective(s) of the subject (Loosemore et al. 2006), suggesting that where there is no
possibility of impact, the subject may fail to recognise the uncertainty as a risk event.
The presentation of findings has been organised under three headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants supported the study theoretical predictions.
Theoretical Issue 3
– Theoretical issue 3: Perceptions are formed from the representative and availability
heuristics (Kahneman 2011; Slovic et al. 2010).
Theoretical issue to be confirmed if the participants’ risk identification on the hypotheti-
cal projects corresponds to their availability heuristics (evident from their responses on the
common construction project risk events).
The related interview questions are 22–24 (see Appendix A). The study data has been
examined for evidence of the influence of grounded heuristics in the risk identification pro-
cesses, using hypothetical construction projects. The interview protocol was designed to
first collect data on the participants’ grounded heuristics before asking them to identify risk
events from hypothetical projects. The rational was to ascertain whether the risk identifi-
cation on the hypothetical projects was done intuitively through reliance on heuristics.
Table B.5 Consistency analysis of the risk perceptions from hypothetical projects A and B (Case 2).
Case Study 2
Force Majeure
Related
Ground
Condition
Related
Yes
Availability and Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
feasibility of
services
connections
(Continued)
Table B.5 (Continued)
Case Study 2
Logistics &
Communication
Related
Delivery of Yes
material to the
site
Financial &
Commercial
Related
Achieving Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
planning
permission &
discharging
conditions
Difficulty in Yes
achieving
statutory design
standards
Political Related
Construction
Related
Operational
Related
238 Appendix B Case 2 Data Presentation
Findings: The responses have been presented in matching patterns of the respondents’
grounded heuristics, against their risk identification on hypothetical projects. The risk
events have also been categorised using the table of construction project risk events
(Table 7.3). The findings have been summarised in Table B.6.
The empirical findings have revealed substantial risk identification correlating with the
study participants’ grounded heuristics, which confirms reliance on the representative and
availability heuristics:
– QS-2: Two out of the four risk identification from hypothetical project A have correspond-
ing grounded heuristics. Two out of the four risk identification from hypothetical project
B have corresponding grounded heuristics.
– PM-2: Three out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Four out of the four risk identification from hypothetical
project B have corresponding grounded heuristics.
– TM-2: Two out of the five risk identification from hypothetical project A have correspond-
ing grounded heuristics. One out of the two risk identification from hypothetical project
B have corresponding grounded heuristics.
– A-2: One out of the four risk identification from hypothetical project A has corresponding
grounded heuristics. One out of the three risk identification from hypothetical project B
has corresponding grounded heuristics.
– E-2: Three out of the four risk identification from hypothetical project A have correspond-
ing grounded heuristics. Two out of the two risk identification from hypothetical project
B have corresponding grounded heuristics.
– CM-2: Two out of the three risk identification from hypothetical project A has corre-
sponding grounded heuristics. Two out of the three risk identification from hypothetical
project B have corresponding grounded heuristics.
Theoretical Issue 4
Ground Condition
Related
Contamination of site Yes Yes Yes Yes Yes Yes Yes Yes
Excavation of site Yes
Previous landuse to Yes
establish ground
conditions
Ecological issues Yes
Ground investigations Yes
Unexpected ground Yes
conditions
Yes Yes No No No No Yes
Design & Quality
Related
Getting the design right Yes
at the concept stage
Availability and Yes Yes Yes Yes Yes Yes Yes
feasibility of service
connections
(Continued)
Table B.6 (Continued)
Case Study 2
QS-2 PM-2 TM-2 A-2 E-2 CM-2
Questions & Responses 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B
Co ordination of Yes
No No Yes Yes No
Financial &
Commercial Related
Procurement of labour Yes Yes Yes Yes
and materials
Completing within Yes Yes Yes Yes Yes Yes Yes Yes
budget
Marketing of completed Yes Yes
properties
Insolvency of suppliers Yes
Yes Yes Yes Yes No Yes Yes No
Legal and Statutory
Approval Related
Achieving planning Yes Yes Yes Yes Yes Yes Yes Yes
permissions &
discharge of conditions
Ownership of land Yes
Third party issues- Yes Yes Yes Yes
partywall, etc
Statutory approvals Yes
including building
regulations
No No Yes Yes No No No Yes Yes
Political Related
(Continued)
Table B.6 (Continued)
Case Study 2
QS-2 PM-2 TM-2 A-2 E-2 CM-2
Questions & Responses 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B
Construction & Site
Related
Safety Yes Yes Yes Yes Yes Yes
Planning and Yes Yes Yes Yes Yes
programming of the
construction works
Yes Yes Yes Yes Yes
Operational Related
2 of 4 2 of 4 3 of 4 4 of 4 2 of 5 1 of 2 1 of 4 1 of 3 3 of 4 2 of 2 2 of 3 2 of 3
B.2 Research Proposition 2: Findings 243
additional information have been termed exposed heuristics, denoting that they have been
generated based on exposure to additional information which has become assimilated
within the pool of heuristics applied in their intuitive evaluations.
Findings: The responses have been presented in matching patterns of the respondents’
initial risk identification on the hypothetical projects, against their subsequent risk iden-
tification on the same hypothetical projects following exposure to additional assessment
information. The risk events have also been categorised using the table of construction
project risk events (Table 7.3). The findings have been summarised in Table B.7.
The empirical evidence has revealed changes in the initial risk identification profiles after
the study participants were exposed to additional assessment information, which confirms
the vulnerability of perception to being manipulated. The changes in risk identification
profiles were observed in the responses from all the participants. As an illustration, the
technical manager whose initial risk appraisal concentrated on health and safety events,
subsequently expanded his responses to include other risk categories after being exposed
to the schedule of construction risk sources. The changes occurred even though the project
settings have not been altered.
The changes in risk identification profiles were observed in the responses from all the
participants:
– QS-2: One out of the four risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. Two out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– PM-2: Two out of the five risk assessment categories on hypothetical project A were
different from the initial assessment. Two out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– TM-2: One out of the four risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. Four out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– A-2: One out of the three risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. Two out of the three risk assessment categories on
hypothetical project B were different from the initial assessment.
– E-2: Two out of the four risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. One out of the two risk assessment categories on
hypothetical project B was different from the initial assessment.
– CM-2: Two out of the five risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Three out of the four risk assessment categories on
hypothetical project B were different from the initial assessment.
Theoretical Issue 5
– Theoretical issue 5: An uncertainty becomes a risk event to a subject if its occurrence can
potentially affect the objective(s) of the subject (Loosemore et al. 2006), suggesting that
where there is no possibility of impact, the subject may fail to recognise the uncertainty
as a risk event.
Theoretical issue is to be confirmed if the participants fail to identify risk events outside
their specialist background (associated with lack of related grounded heuristics).
Table B.7 Analysis of the influence of additional information (exposed heuristics) on perception formation (Case 2).
Case Study 2
QS-2 PM-2 TM-2 A-2 E-2 CM-2
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Force Majeure Related
Landslips Yes Yes Yes
Earthquake Yes Yes
Flood destructions Yes
Yes Yes Yes Yes Yes
Ground Condition Related
Contamination of site Yes Yes Yes Yes Yes Yes Yes Yes
Excavation of site Yes
Polution from adjoining Yes Yes
sites
Ecological issues Yes Yes Yes Yes Yes
Ground investigations Yes
Site investigations Yes
No Yes Yes No Yes No No Yes No
Design & Quality Related
Availability and feasibility Yes Yes Yes Yes Yes Yes
of service connections
Achieving scheme design Yes Yes
standards
New untested technology Yes
Right of Light Yes
Design constraints/ Yes Yes Yes Yes Yes Yes
opportunities resulting
from surveys
Accuracy and Yes
appriopriateness of design
specification
Accuracy of Yes
investigations to feed into
design
Coordinating design Yes Yes
with construction
method
Yes No Yes No Yes No No No No Yes
Logistics &
Communication Related
Site access Yes Yes Yes Yes Yes
Site setup and Yes
organisation of
construction plants
Delivery of material to Yes Yes Yes Yes Yes
(Continued)
TableB.7 (Continued)
Case Study 2
QS-2 P M-2 TM-2 A-2
Questions & Responses E-2 C M-2
Legal and Statutory 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Approval Related
Achieving planning Yes Yes Yes Yes Yes Yes Yes Yes Yes
permissions & discharge of
conditions
Third party issues- Yes Yes Yes Yes Yes
partywall, etc
Statutory approvals Yes Yes Yes
including building
regulations
No No No No No
Political Related
Differences between Yes
regional and national
laws and regulations
Differences in stakeholder Yes
expectations
Yes Yes
Construction & Site Related
Safety Yes Yes Yes
Planning and Yes Yes Yes Yes
programming of the
construction works
Suitability of construction Yes Yes
method
No No Yes No
Operational Related
Ensuring fitness for purpose Yes
Yes
1 of 4 2 of 5 2 of 5 2 of 5 1 of 4 4 of 5 1 of 3 2 of 3 2 of 4 1 of 2 2 of 5 3 of 4
B.2 Research Proposition 2: Findings 247
The related interview question is 24 (see Appendix A). The study data has been exam-
ined to see if the participants’ failure in identifying certain risk events can be attributed to
their lack of related grounded heuristics outside their specialist background. The interview
protocol was designed to solicit the participants’ risk perceptions on hypothetical project
settings using their grounded heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
risk identification on the hypothetical project settings, against the project cumulative risk
identification profiles, to reveal the items missed by each participant. The risk events have
also been categorised using the table of construction project risk events (Table 7.3). The
findings have been summarised in Tables B.8–B.10.
The data from the two hypothetical project settings has been further analysed for consis-
tency in the participants’ missed risk events.
The empirical evidence and consistency examination have revealed that exclusion of
a specialist role from the risk identification exercise would have resulted in some of the
risk events being missed. The individual responses have revealed some risk events being
perceived by singular specialist roles. The contracts manager with his 36 years extensive
experience was the only participant to have perceived right of light as a possible risk to hypo-
thetical project A. The project architect with his six years architectural experiences in roles
including coordination of planning validation site surveys, likewise, was the only partici-
pant to have perceived ecological constraints as possible risk to hypothetical project B. This
suggest that exclusion of the contract manager and architect from the risk identification
exercise may have resulted in the missing of the associated risk events, with consequential
impact on the ensuing risk management response and project performance, assuming the
hypothetical projects and risk events to be real.
Empirical evidence from the direct observations also revealed that the research partic-
ipants were generally calm and relaxed during the interrogations relating to Theoretical
Issues 3–5. Their responses were spontaneous and effortless, suggesting intuitive decision
processing.
The empirical evidence from Theoretical Issues 3–5 have provided substantial support
for research proposition 2. The findings have revealed substantial correlation between
the research participants’ risk perceptions on hypothetical project settings and their
initial grounded heuristics on hypothetical construction project risk events, which suggest
intuitive decision processing based on the representative and availability heuristics. All
the study participants altered their risk evaluation after they were exposed to additional
assessment information in the form of a schedule of construction risk sources, which also
confirms the ease by which perceptions can be manipulated. Analysis of the individual
participants’ responses against the cumulative case responses has also revealed individual
missed risk events outside the respective specialist backgrounds, thereby validating the
research proposition of restricted risk management practices impairing robust risk identi-
fication and decision-making, especially the risk categories associated with the specialist
background of the excluded roles.
Table B.8 Effect of limited heuristics (excluding some project team members) on project risk identification – Hypothetical Project Scenario A.
Case Study 2
Ground
Condition
Related
Demolition & Yes Yes
excavation
Site Yes Yes Yes
contaminations
Missed Missed Missed
Design & Quality
Related
Achievement of Yes Yes Yes
scheme design
standards
Availability and Yes Yes Yes Yes
feasibility of
services
connections
Rights of light Yes Yes
Missed Missed Missed
Logistics &
Communication
Related
Access and egress Yes Yes Yes
into the site
Site setup and Yes Yes
organisation
Missed Missed Missed
Financial &
Commercial
Related
Completing Yes Yes Yes Yes Yes
within budget
Missed Missed
Legal &
Statutory
Approval Related
Achievingn Yes Yes Yes Yes Yes
planning
permission &
discharging
conditions
Achieving third Yes Yes
party approval
from train/ tube
operators
Partywall issues Yes Yes Yes Yes Yes
resulting from
adjoining
properties
(Continued)
Table B.8 (Continued)
Case Study 2
Construction
Related
Programming of Yes Yes Yes
construction
works to achieve
completion date
Safety on site Yes Yes
Missed Missed Missed
Operational
Related
Ground
Condition
Related
Ground Yes Yes Yes Yes
contamination
Site Yes Yes
investigations
Ecological Yes Yes
Missed Missed
Design & Qulity
Related
Availability and Yes Yes Yes Yes
feasibility of
services
connections
Missed Missed Missed
(Continued)
Table B.9 (Continued)
Case Study 2
QS-2 PM-2 TM-2 A-2 E-2 CM-2
Total Risk Project Individual Individual Individual Individual Individual
Perceptions risk Individual Missed- risk Missed- risk Missed- risk Missed- risk Missed- risk Missed-
of 24B events risk events project events project events project events project events project events project
Logistics &
Communication
Related
Access to the site Yes Yes
Delivery of Yes Yes
material to the site
Attracting sub Yes Yes
contractors/
labour to the
project
Missed Missed Missed
Financial &
Commercial
Related
Procuring Yes Yes Yes
subcontractors
and labour
Completing Yes Yes
within budget
Marketing of Yes Yes Yes
completed
properties
Missed Missed
Legal &
Statutory
Approval Related
Difficulty in Yes Yes
achieving
statutory design
standards
Achieving Yes Yes Yes
planning
permission &
discharging
conditions
Missed Missed Missed
Political Related
Construction
Related
Programming Yes Yes
construction
works to achieve
project
completion date
Safety on site Yes Yes
Missed Missed Missed Missed
(Continued)
Table B.9 (Continued)
Case Study 2
QS-2 PM-2 TM-2 A-2 E-2 CM-2
Total Risk Project Individual Individual Individual Individual Individual
Perceptions risk Individual Missed- risk Missed- risk Missed- risk Missed- risk Missed- risk Missed-
of 24B events risk events project events project events project events project events project events project
Operational
Related
GCR, ground condition related; DQR, design and quality related; LCR, logistics and communication
related; FCR, financial and communication related; LSAR, legal and statutory approval related;
CR, construction related.
The third research proposition advances that statistical and probability risk data presenta-
tion and analysis are less likely to be understood and applied in intuitive risk management
systems compared to the use of qualitative data presentation formats and subjective assess-
ment techniques. The underlining theoretical issues are as follows:
– 6. Research findings have identified construction risk management to be dominated by
people without formal training in risk management systems using intuition and heuris-
tics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha
2010).
– 7. Psychometric research findings suggest that most people relate better to narratives,
models, illustrations, and anecdotal simulations better than statistical data (Slovic et al.
2010).
– 8. Most people do not trust, understand, or use statistics and probability estimates in their
risk management decision-making (March and Shapira 1987; Akintoye and Macleod
1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
– 9. Most people relate better to qualitative assessments than quantitative assessments
(Slovic et al. 2010).
– 10. Where a potential outcome of a future event evokes a strong affective memory, the
assessor may be insensitive to the probability predictions on the event (Slovic et al. 2010).
The presentation of findings has been organised under five headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants have supported the study’s theoretical predictions on decision processing within the
construction risk management decision-making subsystem.
256 Appendix B Case 2 Data Presentation
Theoretical Issue 6
– Theoretical issue 6: Research findings have identified construction risk management to
be dominated by people without formal training in risk management systems using intu-
ition and heuristics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga
and Kuotcha 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for intuitive
risk management approaches over rational tools and techniques.
The related interview questions are 7–11 (see Appendix A). The study data has been
examined for evidence on the level of knowledge and application of rational techniques
within the construction risk management decision-making subsystems. The interview pro-
tocol was designed to stimulate intuitive thinking and decision processing by allowing the
participants limited time to respond to the questions.
Findings: The findings from case study 2 have been summarised in Figures B.1–B.9 and
Table B.11.
Interview Question 7 Please indicate your level of education/training in project risk man-
agement tools and practices?
a. No formal education or training in project risk management (………)
b. On the job informal experience (………)
c. Continuing professional developments (CPDs) and seminars on project risk manage-
ment (………)
d. Short duration formal course (less than six months) (………)
e. University diploma/degree in Project Risk Management – e.g. BSc Construction Risk
Management, MSc Risk Management (………)
Please specify title of degree/diploma…………………………………………………………
Findings: The responses have been presented in Figure B.1.
University diploma/degree 0%
Interview Question 8 Which of the following risk management techniques have you heard
of? (Please indicate with a tick)
Risk register ………… . . . .. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) …………… Risk breakdown analysis ………… . . . ..
Worse/best case scenario ……………… Monte Carlo simulation/analysis ………… . . . .
Brainstorming………… . . . .. Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ……… . . . .. Other (please specify) …………………………
Findings: The responses against the risk management techniques have been presented in
Figure B.2. Figure B.3 also categorises the responses into intuitive and rational techniques.
Others 0%
Cultural beliefs – Intuitive 7.84%
Failure mode and effect analysis – Rational 2%
Monte Carlo simulation/analysis – Rational 3.92%
Risk breakdown analysis – Rational 7.84%
Personal feelings and values – Intuitive 12%
Probability matrix – Rational 7.84%
Gut feeling – Intuitive 11.76%
Fish bone – Rational 2%
Brainstorming – Intuitive 11.76%
Worse/best case scenario – Intuitive 9.80%
Personal experiences (heuristics) – Intuitive 11.76%
Fault tree – Rational 2%
Risk register – Intuitive 9.80%
Figure B.2 Risk management techniques known by the participants (all responses).
Rational,
25.49%
Intuitive,
74.51%
Figure B.3 Risk management techniques known by the participants (categorisation into intuitive
and rational).
258 Appendix B Case 2 Data Presentation
Interview Question 9 Which of the following risk management techniques have you
applied on this project? (Please indicate with a tick)
Risk register ………… . . . .. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) …………… Risk breakdown analysis ………… . . . ..
Worse/best-case scenario ……………… Monte Carlo simulation/analysis ………… . . . .
Brainstorming………… . . . .. Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ……… . . . . Other (please specify) ……………………………
Findings: The responses against the applied risk management techniques have been
presented in Figure B.4. Figure B.5 categorises the responses into intuitive and rational
techniques.
Observation and analysis of plans – Intuitive 3%
Cultural beliefs – Intuitive 2.86%
Failure mode and effect analysis – Rational 0%
Monte Carlo simulation/ analysis – Rational 0%
Risk breakdown analysis – Rational 5.71%
Personal feelings and values – Intuitive 17.14%
Probability matrix – Rational 3%
Gut feeling – Intuitive 11.43%
Fish bone – Rational 0%
Brainstorming – Intuitive 14.29%
Worse/ best case scenario – Intuitive 11.43%
Personal experiences (heuristics) – Intuitive 17.14%
Fault tree – Rational 0%
Risk register – Intuitive 14.29%
0.00% 2.00% 4.00% 4.00% 8.00% 10.00% 12.00% 14.00% 16.00% 18.00%
Intuitive,
91.43%
Figure B.5 Risk management techniques applied (categorisation into intuitive and rational).
B.3 Research Proposition 3: Findings 259
Interview Question 10 On a scale of 1–3 please rank the top three (3) techniques which
you would prefer to use on your future risk management practices? You may select from
the following list – Risk register, Probability matrix, Fault tree, Personal feelings and values,
Personal experiences (heuristics), Risk breakdown analysis, Worse/best-case scenario, Monte
Carlo simulation/analysis, Brainstorming, Failure mode and effect analysis, Fish bone, Cul-
tural believes, Gut feeling (please note: A scale of 1 represents the highest preference, and a
scale of 3 represents the lowest preference).
1st ranked technique ………………………………………………………………………
2nd ranked technique ………………………………………………………………………
3rd ranked technique ………………………………………….……………………………
Findings: The responses against the preferred risk management techniques have been
presented in Figure B.6. Figure B.7 categorises the responses into intuitive and rational
techniques.
0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 30.00% 35.00% 40.00% 45.00%
Rational, 7.41%
Intuitive,
92.59%
Figure B.7 Risk management techniques preferred (categorisation into intuitive and rational).
260 Appendix B Case 2 Data Presentation
A trend analysis of the responses on the risk management techniques known, applied,
and preferred has been carried out to see if there are any pattern in the responses that could
shed light on the rationale behind the present risk management practices. The evidence has
been presented in Figures B.8 and B.9.
0 5 10 15 20 25
Figure B.8 Trend analysis of the risk management techniques known, applied, and preferred.
90.00%
80.00% 74.51%
70.00%
60.00%
50.00%
40.00%
25.49%
30.00%
20.00%
8.57% 7.41%
10.00%
0.00%
Known Applied Preferred
Intuitive 74.51% 91.43% 92.59%
Rational 25.49% 8.57% 7.41%
Figure B.9 Trend analysis of the risk management techniques known, applied, and preferred
(categorisation into intuitive and rational).
B.3 Research Proposition 3: Findings 261
The evidence from the trend analysis has revealed high responses for intuitive techniques
across the three assessment variables, which suggest that the risk management practices of
the research participants within the short term will continue to utilise intuitive thinking and
decision-making. The individual participants’ responses have confirmed sole reliance on
intuitive technique for all the specialist roles, apart from the project manager and technical
manager. The contracts manager who was also the project director confirmed that most
private developers do not follow structured risk management systems. His observation in
the course of 36 years industry experience has been the use of internal subsidiary contractor
to complete a risk register at project commencement to satisfy corporate due diligence, with
subsequent reliance on personal experiences and gut feelings to manage day-to-day project
risk events.
The limited knowledge and application of rational risk management techniques among
the case study participants may be partly attributed to the low level of training and educa-
tion in project risk management tools and techniques presented in Figure B.1.
The empirical evidence on the factors influencing the participants’ preference for the
various risk management techniques has been addressed in Interview Question 11.
Interview Question 11 What are the factors influencing your preference of the above risk
management technique?
Theoretical Issue 7
– Theoretical issue 7: Psychometric research findings suggest that most people relate better
to narratives, models, illustrations, and anecdotal simulations better than statistical data
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for quali-
tative data presentation formats, over statistical data, and quantitative models.
The related interview questions are 12–14 (see Appendix A). The study data has
been examined for evidence on preference for construction risk management data pre-
sentation format. The interview protocol was designed to stimulate intuitive thinking
and decision processing by allowing the participants limited time to respond to the
questions.
Findings: The findings from case study 2 have been summarised in Figures B.10–B.17.
262 Appendix B Case 2 Data Presentation
Case study 2
Risk register Good for risk allocation and as an audit trail for 1
checking risk management responses
Enable all project risk to be documented in a single 1
document for monitoring and treatment
Simple way of recording and presenting project risk 2
from project inception
Familiar method of risk identification and treatment 1
Personal experience Opportunity to draw from lessons learnt on previous 3
projects
Instinctive way of identifying risk 1
Brainstorming Opportunity to utilise knowledge from all the project 1
team rather than just one person
Personal feeling and Opportunity to utilise the wealth of professional 1
values experiences in project risk management
Risk breakdown analysis Enable risk to be broken down for better clarity 1
Monte Carlo An opportunity to transfer build-up experience into 1
simulation/analysis project risk identification and treatment
Worse/best case scenario Instinctive way of identifying risk 1
Opportunity to analysis risk associated with different 2
project scenarios
Observation and analysis Practical way of identifying design risk 1
of plans
Gut feeling Use your experience in risk identification and 1
treatment
Interview Question 12 Which of the following forms of risk data presentations do you
know? (Please tick all the applicable)
Findings: The responses against the risk management data presentation formats have
been presented in Figure B.10. Figure B.11 also categorises the responses into qualitative
and quantitative presentation formats.
B.3 Research Proposition 3: Findings 263
63.16%
Qualitative Quantitative
Interview Question 13 Which of the following forms of risk data presentations have you
applied on this project? (Please tick all the applicable)
Findings: The responses against the risk management data presentation formats have
been presented in Figure B.12. Figure B.13 also categorises the responses into qualitative
and quantitative presentation formats.
264 Appendix B Case 2 Data Presentation
0.00%2.00%4.00% 6.00%8.00%10.00%12.00%14.00%16.00%18.00%20.00%
60.87%
Qualitative Quantitative
Interview Question 14 On a scale of 1–3, please rank the top three (3) risk data presentation
formats which you would prefer to use on your future risk management practices? You may
select from the following list – Narratives, Probabilities, Qualitative statements, Whole num-
bers, Graphs, Ratios, Fractions, Scenarios, and Simulations (please note: A scale of 1 represents
the highest preference, and a scale of 3 represents the lowest preference).
B.3 Research Proposition 3: Findings 265
Findings: The responses against the risk management data presentation formats have
been presented in Figure B.14. Figure B.15 also categorises the responses into qualitative
and quantitative presentation formats.
Fractions – Quantitative
56.60%
Qualitative Quantitative
266 Appendix B Case 2 Data Presentation
A trend analysis of the responses on the risk management data presentation formats
known, applied, and preferred has been carried out to see if there are any pattern in the
responses that could shed light on the rationale behind the present risk management prac-
tices. The evidence has been presented in Figures B.16 and B.17.
Others
Sketches and mark-ups on drawings –Qualitative 5.00
1.00
1.00
Using different colours to rank severity of 5.00
2.00
risk – Qualitative 1.00
Scenarios – Qualitative 7.00
2.00
4.00
6.00
Ratios – Quantitative 3.00
4.00
Whole numbers – Qualitative Preferred
4.00
6.00 Applied
Probabilities – Quantitative 4.00
5.00 Known
5.00
Simulations – Qualitative 2.00
4.00
Fractions – Quantitative
11.00
Graphs – Quantitative 2.00
5.00
4.00
Qualitative statement – Qualitative 3.00
5.00
4.00
Narrative – Qualitative 4.00
6.00
0.00 2.00 4.00 6.00 8.00 10.00 12.00
Figure B.16 Trend analysis of the risk management data presentation formats known, applied, and
preferred.
70.00% 63.16%
60.87%
60.00% 56.60%
50.00% 43.40%
39.13%
36.84%
40.00%
30.00%
20.00%
10.00%
0.00%
Known Applied Preferred
Qualitative 63.16% 60.87% 56.60%
Quantitative 36.84% 39.13% 43.40%
Figure B.17 Trend analysis of the risk management data presentation formats known, applied, and
preferred (categorisation into qualitative and quantitative).
B.3 Research Proposition 3: Findings 267
The evidence from the trend analysis has revealed comparatively higher responses
for qualitative data presentation formats across the three assessment variables, which
suggest prolonged existence for the research participants’ current risk management
practice of qualitative data processing. The individual responses have also revealed
preference for qualitative risk data presentation for all the specialist roles. The quantity
surveyor asserted that pictorial risk data presentation formats including the use of differ-
ent colours and illustrations for risk classification, enhances visual appreciation of the
risk event.
Notwithstanding the above, the trend analysis has revealed disparities between the
proportions of applied systems of risk data presentation formats (60.87% qualitative against
39.84% quantitative) versus the systems of applied risk management techniques presented
under Theoretical Issue 6 (94.43% intuitive against 8.75% rational). There is also a pattern
of progressive increase in the application and preference for quantitative data presentation
formats, which also appears theoretically incompatible with the trend analysis of compar-
ative reduction in application and preference for rational risk management techniques,
presented under Theoretical Issue 6. The previous theoretical discussion on the need for
complementarity between a decision processing system and its data presentation format
(Kahneman 2011) suggests the likelihood of psychological conflicts and reduced accuracy
in the risk analysis output of the case study participants.
Theoretical Issue 8
– Theoretical issue 8: Most people do not trust, understand, or use statistics and probability
estimates in their risk management decision-making (March and Shapira 1987; Akintoye
and Macleod 1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
Theoretical issue to be confirmed if the empirical evidence reveals low level of knowledge
and application of statistics and probability.
The related interview questions are 15–18 (see Appendix A). The study data has been
examined for evidence on the participants’ level of knowledge and understanding of statis-
tics and probability assessment. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions intuitively without consulting any reference source.
Findings: The findings from case study 2 have been summarised in Table B.12 and
Figures B.18–B.20.
Interview Question 15 Which of the following statistical symbols are you familiar with?
Please indicate by selecting from the following list: summation, probability density func-
tion, probability function, sample mean, population mean, median, variance, standard devi-
ation, sample standard deviation, and don’t know.
268 Appendix B Case 2 Data Presentation
Symbol Name
P(A)
S
∑
Case study 2
Interview Question 16–18: Please use the hypothetical risk evaluations represented in the
table below to answer questions 16–18.
Probability of Probability of
Risk event occurrence impact
A 1/7 30/35
B 97.9% 84%
C 1 : 9743 4:5
100.00%
Interview Question 17 Which of the above risk events is certain not to occur?
Risk event A ……………………………………… . . . ..
Risk event B ……………… . . . ..
Risk event C ……………………………………… . . . ..
All (risk events A, B, and C) ………………………
None of the risk events …………………… . . . ..
Don’t know ……………………………………… . . . ..
Findings: The percentage of correct responses has been presented in Figure B.19.
83.33%
Interview Question 18 Which of the above risk events is certain to have the highest impact?
Risk event A ……………………………………… . . . ..
Risk event B ……………… . . . ..
Risk event C ……………………………………… . . . ..
All (risk events A, B, and C) …………………… . . . .
None of the risk events ………………………… . . . .
Don’t know ……………………………………………
270 Appendix B Case 2 Data Presentation
Findings: The percentage of correct responses has been presented in Figure B.20.
100.00%
The empirical evidence has revealed significant low level of understanding of statistical
symbols and probability assessments among the case study participants:
– 5 out of 24 correct identification of statistical symbols which represents 20.83% (see
Table B.12). The project manager asserted that most construction professionals do
not use statistics in their regular activities. He attributed this to a general perception
that statistical data presentation is usually applied when professionals want to deceive
laymen.
– 1 out of 18 correct responses for the probability assessments which represents 5.56%.
A significant empirical observation is the wrong responses from the project manager,
technical manager, engineer, and contracts manager whose other research responses
confirms sole reliance on intuitive risk analysis of data formats including quantitative
probability. Another significant empirical discovery is the erroneous linkage of higher
probability prediction to certainty of occurrence of an event, which may have resulted
from a general lack of understanding of the probability concept.
Theoretical Issue 9
– Theoretical issue 9: Most people relate better to qualitative assessments than quantitative
assessments (Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals high accuracy for risk
analysis expressed in whole numbers, qualitative statements, and scenarios, compared to
risk analysis expressed in percentages, decimals, and ratios.
The related interview questions are 19 and 20 (see Appendix A). The study data has
been examined for evidence on the participants’ level of knowledge and understanding of
qualitative and quantitative assessments. The interview protocol was designed to stimu-
late intuitive thinking and decision processing by allowing the participants limited time to
respond to the questions intuitively without using calculators.
Findings: The findings from case study 2 have been summarised in Figures B.21 and B.22
and Table B.13.
B.3 Research Proposition 3: Findings 271
Interview Question 19 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. There is a 75.2% probability of half of the site being contaminated (………)
B. There is a 1 : 321 probability of the site being contaminated (………)
C. There is a 0.381 probability of the site being contaminated (………)
Findings: The percentage of correct responses has been presented in Figure B.21.
100.00%
Interview Question 20 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. The probability of projects over running their initial budget is 3 out of 10 (………)
B. out of every projects overrun their initial budget (………)
C. After reviewing the financial records of the four completed construction projects, only
project A recorded a final cost higher than the initial budget estimate (………)
Findings: The percentage of correct responses has been presented in Figure B.22.
66.67%
The empirical evidence has revealed higher percentage of accuracy for the qualitative
probability equations compared to the quantitative probability equations:
– Qualitative probability equation: Four out of six correct responses representing 66.67%.
– Quantitative probability equation: Zero out of six correct response representing 0%.
Another significant evidence is the composition of the wrong responses. Five out of
the six wrong respondents selected the probability equation expressed in percentages,
representing 83.33% of the wrong responses. Further questioning on the rationale
behind the preference for the ‘percentage equation’ format revealed the following
responses:
a. Percentages are what we are familiar with from our primary and secondary educa-
tion – two responses.
b. Percentages are what we use in our everyday working life – one response.
c. Percentages are easy to understand – two responses.
The above responses suggest that the participants who selected the percentage equation
were simply relying on their availability heuristics to intuitively generate the answers.
There were two further supplementary questions to solicit empirical evidence on the par-
ticipants’ experience with answering the two forms of probability equations. The findings
have been presented in Table B.13.
Table B.13 Participants’ experience with answering the quantitative and qualitative probability
equations.
The empirical evidence revealed balanced responses for both quantitative and qualita-
tive data presentation formats. The indifference in the study participants’ experience and
the preceding empirical observation of low accuracy in intuitive risk analysis of quanti-
tative probability equations may have resulted from the psychological conflict of mixing
tools and techniques from the different systems of thinking, discussed under Theoretical
Issue 7.
Theoretical Issue 10
– Theoretical issue 10: Where a potential outcome of a future event evokes a strong affec-
tive memory, the assessor may be insensitive to the probability predictions on the event
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals insensitivity
to different ranges of hypothetical probability predictions of asbestos contamination
risk.
B.3 Research Proposition 3: Findings 273
The related interview questions are 21i–21iii (see Appendix A). The study data has been
examined for evidence on the participants’ responses to the different range of hypothetical
asbestos contamination risk. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions.
Findings: The findings from case study 2 have been summarised in Table B.14.
Table B.14 The impact of probability predictions on events which evokes strong affective
memories.
Probability of Probability of
Hypothetical occurrence based impact based on Appropriate Responses/
setting on initial survey initial survey risk response percentages
Interview Question 21 Asbestos related illness has been a major cause of construction
occupational death. According to the Health and Safety Executive in the United Kingdom,
there has been a recent surge in asbestos related illness and deaths (see the table and
pictures below).
Asbestos related deaths
Asbestos-related
cancer mesothelioma
Year deaths Asbestosis death
1968 153
2009 2321 411
2010 2347 412
2011 2291 429
2012 2535 464
Source: Shutterstock.com
B.3 Research Proposition 3: Findings 275
21ii What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
(Please select your response from any of the options given under 21i.)
21iii What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
(Please select your response from any of the options given under 21i.)
Findings: The study participants’ risk responses under the different probability predic-
tions have been presented in Table B.14.
The empirical evidence has revealed significant insensitivity to the different ranges of
hypothetical probability predictions on asbestos contamination. 83.33% of the participants
consistently responded that they would appoint a licensed contractor to carry out further
asbestos surveys to identify and remove asbestos contaminated materials before they com-
mence site construction works. The different levels of probability risk predictions could
not alter their responses, not even for the very low probability prediction of non-malignant
contamination within a small specified area. The contracts manager who confirmed los-
ing his father from mesothelioma in the preceding year was very emotional during the
interrogations. He indicated that his risk response to asbestos contamination will always
be avoidance irrespective of the probability estimate. His zero tolerance to asbestos risk
may have been precipitated by an inducement of a strong affective feeling associated with
his previous bitter experience of observing his father dying from asbestos cancer.
The empirical evidence from the direct observations also revealed some level of emotional
discomfort for some of the participants during the interrogations, especially the statistical
and probability assessments (Theoretical Issue 8) and the quantitative equations (Theo-
retical Issue 9). Some of the participants even inquired if they could search the statistical
symbols on the internet and process the quantitative equations with calculators. When this
B.3 Research Proposition 3: Findings 277
was denied, the frustration became evident in their body language, confirming the psy-
chological conflict they were facing in attempting intuitive processing of quantitative data
formats.
In summary, the empirical evidence from Theoretical Issues 6–10 has provided substan-
tial support for research proposition 3. The findings have revealed intuitive dominated risk
management practices and limited training and education in formal risk management sys-
tems. There is also evidence of disparities between the systems of applied and preferred risk
management techniques versus the applied and preferred risk data presentation formats,
with evidence of intuitive processing of theoretically incompatible quantitative probability
data. The study participants have demonstrated low proficiency in statistical symbols and
probability judgement. They have also exhibited reasonable accuracy for intuitive evalua-
tion of qualitative data, a complete failure for intuitive evaluation of quantitative data, and
insensitivity to probability predictions of emotive events.
279
Appendix C
Appendix C presents the empirical evidence from case study 3. The findings have been pre-
sented under three subsections corresponding to the three research propositions developed
in Section 6.2. The findings for the propositions have been further organised under sepa-
rate headings relating to the underlining theoretical issues. The presentation has adopted
the question and answer approach. The interview questions relating to each theoretical
issue have been described in Appendix A.
Theoretical Issue 1
– Theoretical issue 1: Systems differentiation also generates differences in risk perceptions
among the system components – different specialist roles (Walker 2007, 2015; Kahneman
2011; Slovic et al. 2010).
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
280 Appendix C Case 3 Data Presentation
Interview Question 23 Drawing from your professional experiences, what will you say are
the factors likely to affect project success (risk events) at the construction phase (during
actual construction work on site)?
Findings: The responses have been presented in Table C.2 using the same format as the
previous question Interview Question 22.
The empirical evidence on the typical construction phase risk events at the construction
phase has also revealed differentiation within the generated risk perceptions, with a pattern
of categorisation corresponding with the study participants’ specialist backgrounds. The
differentiated builder specialist subgroup comprising the project manager and contracts
manager generated risk perceptions in the following order of ranking: ground condition
related (first), legal and statutory approval related (second), and financial and commercial
related (third). The designers comprising the engineer and architect also generated risk per-
ceptions in the following order of ranking: financial and commercial related (first), ground
condition and construction related (second), and design and quality related (third). The
quantity surveyor’s generated risk perceptions on the other hand were in the following
order of ranking: financial and commercial related (first), construction related (second),
and ground condition related (third). The client generated risk perceptions were all con-
struction related. Upon further interrogations, the client confirmed previously working as
a site manager before embarking on a career change into his present role.
Theoretical Issue 2
– Theoretical issue 2: The differences in objectives of differentiated subsystems (Carmichael
2006; Loosemore et al. 2006) suggest that the impact of environmental forces will result
in differences in affective heuristics, and therefore differences in perceptions within the
system components – different specialist roles.
Theoretical issue to be confirmed if the empirical evidence reveals similarities in the
respondents’ risk perceptions on hypothetical project settings and their specialist back-
ground.
The related interview question is numbered Interview Question 24 (see Appendix A).
The rationale behind this question was to solicit the participants’ risk perceptions on two
hypothetical construction project settings.
Interview Question 24 The following pictures depict hypothetical sites (edged in bold
front) being considered for residential development. The project is at the inception stage
with the following key project objectives:
– Design: 50 units to meet minimum building regulation requirements, lifetime homes,
secured by design, minimum 40% carbon reduction.
– Budget: Construction cost not to exceed £15 million.
– Programme: To be completed within two years.
C.1 Research Proposition 1: Findings 283
Hypothetical site A: Brown field development within high density urban area
Findings: The responses outline the participants’ risk perceptions on the hypothetical
projects. The risk events have been categorised using the table of construction project risk
events (Table 7.3). The findings have been summarised in Tables C.3–C.5.
Case study 3
Cumulative Cumulative
Risk perception PM-3 CM-3 for builders A-3 E-3 for engineers QS-3 CDM-3
Force majeure related
Ground condition
related
Site contaminations Yes Yes Yes
Archeological findings Yes Yes
Demolition – asbestos Yes
Logistics and
communication related
Access and egress into Yes Yes
the site
Site setup and Yes Yes
organisation
Access for delivery Yes Yes Yes Yes
vehicles
Availability of specialist Yes Yes
labour
Financial and
commercial related
Completing within Yes Yes
budget
Subcontractor Yes
procurement
(Continued)
286 Appendix C Case 3 Data Presentation
Case study 3
Cumulative Cumulative
Risk perception PM-3 CM-3 for builders A-3 E-3 for engineers QS-3 CDM-3
Political related
Construction related
Programming of Yes
construction works to
achieve completion
date
Working at height Yes
Operational related
Ensuring building Yes
meets the needs of
prospective tenants
(fitness for purpose)
There is further comparison of the data from the two hypothetical projects to check for
consistency and also to examine the pattern of the generated risk perceptions for evidence of
differentiation along the lines of the research participants’ respective specialist background.
The consistency analysis was also meant for filtering out the other extraneous risk forma-
tion variables which are unrelated to specialist role affective heuristics.
The empirical findings have revealed evidence of differentiation within the generated
risk perceptions from the two hypothetical projects, with the pattern of categorisation being
consistent with the specialist background of the study participants. The consistency analysis
has also revealed the following:
– Ground condition related risk events were perceived by the project manager and quantity
surveyor.
C.1 Research Proposition 1: Findings 287
Case study 3
Cumulative Cumulative
Risk perception PM-3 CM-3 for builders A-3 E-3 for designers QS-3 CDM-3
Force majeure related
Weather Yes Yes
Fire Yes
Ground condition
related
Ground contamination Yes
Contamination Yes
resulting from previous
landuse
Ecological Yes Yes Yes
Archeological findings Yes Yes
Logistics and
communication related
Access to the site Yes
Attracting sub Yes Yes
contractors/labour to
the project
Financial and
commercial related
Supply chains – Yes Yes
procurement of
materials
(Continued)
288 Appendix C Case 3 Data Presentation
Case study 3
Cumulative Cumulative
Risk perception PM-3 CM-3 for builders A-3 E-3 for designers QS-3 CDM-3
Political related
Construction related
Site security Yes
Foundation solution Yes
Operational related
Getting residents to Yes
occupy completed
properties
– Design and quality related risk events were perceived by the project manager, architect,
client development manager, and engineer.
– Logistics and communication related risk events were perceived by the contracts man-
ager, and quantity surveyor.
– Legal and statutory approval related risk events were perceived by the project manager
and architect.
– Construction related risk event was also perceived by the quantity surveyor.
Operational related risk event was perceived by the client development manager.
Evaluation of the evidence from the consistency analysis from the perspective of the dif-
ferent specialist roles has also revealed the following categorisations consistent with the
differences in specialist background:
– The builders comprising the project manager and the contracts manager have exhibited
risk perceptions associated with ground condition, design and quality, logistics and com-
munication, and legal and statutory related events.
– The designers comprising the architect and the engineer also displayed risk perceptions
relating to design and quality and legal and statutory approval events. Design and quality
risk events were consistent in all of the responses.
Table C.5 Consistency analysis of the risk perceptions from hypothetical projects A & B (Case 3).
Case Study 3
Force Majeure
Related
Fire Yes
Ground
Condition
Related
Demolition- Yes
asbestos
Ground Yes
contamination
Contamination Yes
resulting from
previous landuse
(Continued)
Table C.5 (Continued)
Case Study 3
Availability and Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
feasibility of
services
connections
Logistics &
Communication
Related
Financial &
Commercial
Related
Subcontractor Yes
procurement
Achieving Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
planning
permission &
discharging
conditions
Political Related
(Continued)
Table C.5 (Continued)
Case Study 3
Construction
Related
Working at Yes
height
Foundation Yes
solution
Yes
Operational
Related
Ensuring Yes
building meets
the needs of
prospective
tenants (fitness
for purpose)
Yes
C.2 Research Proposition 2: Findings 293
– The quantity surveyor on the other hand has exhibited risk perceptions relating to logis-
tics and communication, ground condition, and construction events.
– The client has also displayed risk perceptions associated with operational and design and
quality related events.
In summary, the empirical evidence from Theoretical Issues 1 and 2 have supported
research proposition 1. The findings have revealed differentiation in risk perception
categorisations on the typical construction project risk events at the pre-construction;
and actual construction delivery phases, and also for risk assessments under different
project settings, with the pattern of differentiation correlating with the study participants’
specialist background. The realisation of consistent findings under the different project
delivery phases and settings have invariably confirmed the research novel theoretical
proposition of systems differentiation influencing risk perception categorisation within the
construction project delivery system, with the pattern reflecting the differences in affective
heuristics of the assembled specialist roles.
Theoretical Issue 3
– Theoretical issue 3: Perceptions are formed from the representative and availability
heuristics (Kahneman 2011; Slovic et al. 2010).
Theoretical issue to be confirmed if the participants’ risk identification on the hypotheti-
cal project settings corresponds to their availability heuristics (evident from their responses
on the common construction project risk events).
The related Interview Questions are 22–24 (see Appendix A). The study data has been
examined for evidence of the influence of grounded heuristics in the risk identification
processes, using hypothetical construction project settings. The interview protocol was
designed to first collect data on the participants’ grounded heuristics before asking them
to identify risk events from hypothetical projects. The rational was to ascertain whether
the risk identification on the hypothetical project settings was done intuitively through
reliance on heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
grounded heuristics, against their risk identification on hypothetical project settings. The
risk events have also been categorised using the table of construction project risk events
(Table 7.3). The findings have been summarised in Table C.6.
The empirical findings have revealed substantial risk identification correlating with the
participants’ grounded heuristics, which confirms reliance on the representative and avail-
ability heuristics:
– PM-3: Three out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Three out of the four risk identification from hypothetical
project B have corresponding grounded heuristics.
– CM-3: One out of the two risk identification from hypothetical project A has correspond-
ing grounded heuristics. One out of the two risk identification from hypothetical project
B has corresponding grounded heuristics.
– QS-3: Three out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Two out of the three risk identification from hypothetical
project B have corresponding grounded heuristics.
– A-3: One out of the three risk identification from hypothetical project A has correspond-
ing grounded heuristics. One out of the two risk identification from hypothetical project
B has corresponding grounded heuristics.
– CDM-3: Two out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. One out of the four risk identification from hypothetical
project B has corresponding grounded heuristics.
– E-3: One out of the three risk identification from hypothetical project A has correspond-
ing grounded heuristics. One out of the one risk identification from hypothetical project
B has corresponding grounded heuristics.
Theoretical Issue 4
– Theoretical issue 4: The inherent subjectivity of perception makes it manipulative
and less rational. Additional information could change risk perception even where
Table C.6 Analysis of perception being formed from availability and representative heuristics (Case 3).
Case study 3
PM-3 CM-3 QS-3 A-3 CDM-3 E-3
Questions & Responses 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B
Force Majeure Related
Weather condition Yes Yes
Fire Yes
No No
Ground Condition Related
Contamination of site Yes Yes Yes Yes Yes Yes
Excavation of site Yes
Previous landuse to establish Yes
ground conditions
Site constraints, eg congested Yes
site
Ecological issues Yes Yes Yes
Asbestos Yes Yes
Unexpected ground conditions Yes
Unexpected archeological Yes Yes Yes
findings
Yes Yes Yes Yes Yes
Design & Quality Related
Getting the design right at the Yes
concept stage
Availability and feasibility of Yes Yes Yes Yes Yes Yes Yes
service connections
(Continued)
TableC.6 (Continued)
Case study 3
PM-3 CM-3 Q -3S
Questions & Responses A-3 C M-3D E-3
Impact of existing service 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22
Yes & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B
layouts on design
Co ordination of design team Yes
Achieving scheme design Yes Yes
standards
Achieving lifetime home Yes
standards within site
constraints
Parking provisions Yes
Availability of design Yes
information and specification
Client design changes/ Yes
variations
the base situation has not changed – best case scenarios which lead to gaps between
pre-commencement estimates and actual values at completion stage (Flyvbjerg et al.
2003; Alhakami and Slovic 1994; Slovic and Peters 2006).
The changes in risk identification profiles were observed in the responses from all the
participants:
– PM-3: Three out of the five risk assessment categories on hypothetical project A were
different from the initial assessment. One out of the four risk assessment categories on
hypothetical project B was different from the initial assessment.
– CM-3: Two out of the four risk assessment categories on hypothetical project A were
different from the initial assessment. Three out of the four risk assessment categories on
hypothetical project B were different from the initial assessment.
– QS-3: Two out of the four risk assessment categories on hypothetical project A were
different from the initial assessment. Two out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– A-3: Three out of the five risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Three out of the three risk assessment categories on
hypothetical project B were different from the initial assessment.
Table C.7 Analysis of the influence of additional information on perception formation (Case 3).
Case study 3
PM- 3 CM- 3 QS- 3 A- 3 CDM- 3 E- 3
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Force Majeure Related
Weather condition Yes
Fire destruction Yes Yes Yes Yes
Flood destruction Yes Yes Yes
Earthquake Yes
Yes No Yes Yes No Yes
Ground Condition Related
Contamination of site Yes Yes Yes Yes Yes
Polution from adjoining sites Yes Yes
Previous landuse to establish Yes
ground conditions
Ecological issues Yes Yes Yes Yes Yes Yes
Asbestos Yes
Unexpected archeological Yes Yes
findings
No Yes No Yes Yes No Yes Yes
Design & Quality Related
Availability and feasibility of Yes Yes Yes Yes Yes Yes
service connections
Co ordination of design team Yes
Achieving scheme design Yes Yes
standards
(Continued)
Table C.7 (Continued)
Case study 3
PM- 3 CM- 3 QS- 3 A- 3 CDM- 3 E- 3
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Achieving lifetime home Yes
standards within site
constraints
Parking provisions Yes
Availability of design Yes Yes
information and specification
Design constraints and Yes Yes
opportunities resulting from
surveys
Accuracy of investigations to Yes
feed into designs
Difficulty in achieving Yes
required units within site
constraints
Achieving secured by design Yes Yes
Coordinating design with Yes Yes
construction method
No Yes Yes No No No No No
Logistics & Communication
Related
Site access Yes Yes Yes Yes Yes
Site setup and organisation of Yes
construction plants
Site setup, welfare and Yes
deliveries
Delivery of material to the site Yes Yes Yes Yes Yes Yes Yes
Availability of specialist Yes Yes Yes
labour
Access for delivery vehicles Yes
Waste transfer from site Yes
No No Yes No Yes Yes No Yes
Financial & Commercial
Related
Procurement of labour and Yes Yes
material
Completing within budget Yes
Availability of funds Yes Yes Yes
Supply chains management Yes
Constructors and supplier’s Yes
default
Fluctuations in market Yes Yes
demand
Adequacy of insurance Yes Yes
Yes Yes No Yes Yes Yes
Legal and Statutory Approval
Related
Achieving planning Yes Yes Yes Yes Yes Yes
permissions & discharge of
conditions
Third party issues- partywall, Yes Yes Yes Yes
etc
Statutory approvals including Yes
building regulations
No No Yes Yes
Political Related
Differences between regional Yes
and national laws and
regulations
Changes in laws and Yes
regulations
Yes Yes
(Continued)
Table C.7 (Continued)
Case study 3
PM- 3 CM- 3 QS- 3 A- 3 CDM- 3 E- 3
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Construction & Site Related
Planning and programming of Yes Yes Yes
the construction works
Site Security Yes
Construction technique Yes
Working at height Yes
Foundation solution Yes
Suitability of construction Yes Yes Yes Yes
method
Subtotal No Yes No No No Yes Yes
Operational Related
Getting residents to occupy Yes
completed properties
Building to meet the needs of Yes Yes
prospective tenants (fitness for
purpose)
Subtotal Yes
3 of 5 1 of 4 2 of 4 3 of 4 2 of 4 2 of 5 3 of 5 3 of 3 2 of 4 2 of 5 2 of 3 3 of 4
C.2 Research Proposition 2: Findings 303
– CDM-3: Two out of the four risk assessment categories on hypothetical project A were
different from the initial assessment. Two out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– E-3: Two out of the three risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Three out of the four risk assessment categories on
hypothetical project B were different from the initial assessment.
Theoretical Issue 5
– Theoretical issue 5: An uncertainty becomes a risk event to a subject if its occurrence can
potentially affect the objective(s) of the subject (Loosemore et al. 2006), suggesting that
where there is no possibility of impact, the subject may fail to recognise the uncertainty
as a risk event.
Theoretical issue is to be confirmed if the participants fail to identify risk events outside
their specialist background (associated with lack of related grounded heuristics).
The related Interview Question is 24 (see Appendix A). The study data have been exam-
ined to see if the participants’ failure in identifying certain risk events can be attributed to
their lack of related grounded heuristics outside their specialist background. The interview
protocol was designed to solicit the participants’ risk perceptions on hypothetical project
settings using their grounded heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
risk identification on the hypothetical project settings, against the project cumulative risk
identification profiles, to reveal the items missed by each participant. The risk events have
also been categorised using the table of construction project risk events (Table 7.3). The
findings have been summarised in Tables C.8–C.10.
The data from the two hypothetical project settings has been further analysed for consis-
tency in the participants’ missed risk events.
The empirical evidence and consistency examination have confirmed that exclusion of
a specialist role from the risk identification exercise would have resulted in some of the
risk events being missed. The individual responses have revealed some risk events being
perceived by singular specialist roles. The client development manager for instance, was
the only specialist role to have perceived operational related risk events for both hypothet-
ical project settings, which suggest that excluding him from the risk identification exercise
may have resulted in total miss of the operational related risk events with consequential
impact on the ensuing risk management response and project performance, assuming the
hypothetical projects and risk events to be real.
Empirical evidence from the direct observations also revealed that the research partic-
ipants were generally calm and relaxed during the interrogations relating to Theoretical
Issues 3–5. Their responses were spontaneous and effortless, suggesting intuitive decision
processing (Kahneman 2011).
In summary, the empirical evidence from Theoretical Issues 3–5 has provided substantial
support for research proposition 2. The findings have revealed risk perception generation
Table C.8 Effect of limited heuristics (excluding some project team members) on project risk identification – Hypothetical project A.
Case study 3
PM- 3 CM- 3 QS- 3 A- 3 CDM- 3 E- 3
Risk Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Perceptions of risk risk risk risk risk risk risk risk risk risk risk risk risk
24A events events events events events events events events events events events events events
Force Majeure
Related
Ground
Condition
Related
Site Yes Yes Yes
contaminations
Archeological Yes Yes
findings
Demolition- Yes Yes
asbestos
Missed Missed Missed Missed
Design &
Quality Related
Achievement of Yes Yes
scheme design
standards
Availability and Yes Yes Yes Yes
feasibility of
services
connections
Design Yes Yes
coordination
Missed Missed Missed
Logistics &
Communication
Related
Access and Yes Yes Yes
egress into
FMR, force majeure related; GCR, ground condition related; DQR, design and quality related; LCR,
logistics and communication related; FCR, financial and communication related; LSAR, legal and statutory
approval related; PR, political related; CR, construction and site related; OR, operational related.
on hypothetical project settings consistent with the grounded heuristics of the study par-
ticipants, which confirms intuitive decision processing fuelled by the representative and
availability heuristics. The empirical findings have also revealed variations in the study par-
ticipants risk assessment of hypothetical project settings after being exposed to additional
information in the form of a schedule of construction risk sources, which also validates
the susceptibility of perceptions to manipulation. In addition, a review of the study par-
ticipants’ individual risk assessments against the cumulative case risk identification profile
has revealed missed risk events outside their respective specialist backgrounds, thereby val-
idating the research proposition of restricted risk management practices impairing robust
risk identification and decision-making, especially the risk categories associated with the
specialist background of the excluded roles.
The third research proposition advances that statistical and probability risk data presenta-
tion and analysis are less likely to be understood and applied in intuitive risk management
systems compared to the use of qualitative data presentation formats and subjective assess-
ment techniques. The underlining theoretical issues are as follows:
– 6. Research findings have identified construction risk management to be dominated by
people without formal training in risk management systems using intuition and heuris-
tics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga and Kuotcha
2010).
C.3 Research Proposition 3: Findings 313
– 7. Psychometric research findings suggest that most people relate better to narratives,
models, illustrations, and anecdotal simulations better than statistical data (Slovic et al.
2010).
– 8. Most people do not trust, understand, or use statistics and probability estimates in their
risk management decision-making (March and Shapira 1987; Akintoye and Macleod
1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
– 9. Most people relate better to qualitative assessments than quantitative assessments
(Slovic et al. 2010).
– 10. Where a potential outcome of a future event evokes a strong affective memory, the
assessor may be insensitive to the probability predictions on the event (Slovic et al. 2010).
The presentation of findings has been organised under five headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants have supported the study’s theoretical predictions on decision processing within the
construction risk management decision-making subsystem.
Theoretical Issue 6
– Theoretical issue 6: Research findings have identified construction risk management to
be dominated by people without formal training in risk management systems using intu-
ition and heuristics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga
and Kuotcha 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for intuitive
risk management approaches over rational tools and techniques.
The related Interview Questions are 7–11 (see Appendix A). The study data has been
examined for evidence on the level of knowledge, application, and preference for rational
risk management techniques within the construction risk management decision-making
subsystem. The interview protocol was designed to stimulate intuitive thinking and deci-
sion processing by allowing the participants limited time to respond to the questions.
Findings: The findings from case study 3 have been summarised in Figures C.1–C.9 and
Table C.11.
Interview Question 7 Please indicate your level of education/training in project risk man-
agement tools and practices?
a. No formal education or training in project risk management (…….)
b. On the job informal experience (…….)
c. Continuing professional developments (CPDs) and seminars on project risk manage-
ment (…….)
d. Short duration formal course (less than six months) (…….)
e. University diploma/degree in Project Risk Management – e.g. BSc Construction Risk
Management, MSc Risk Management (…….)
Please specify title of degree/diploma…………………………………………………………
Findings: The responses have been presented in Figure C.1.
314 Appendix C Case 3 Data Presentation
University diploma/degree 0%
Interview Question 8 Which of the following risk management techniques have you heard
of? (Please indicate with a tick)
Risk register ……………. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) …………… Risk breakdown analysis …………….
Worse/best case scenario ……………… Monte Carlo simulation/analysis ……………
Brainstorming……………. Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling …………. Other (please specify) …………………………
Findings: The responses against the risk management techniques have been presented in
Figure C.2. Figure C.3 also categorises the responses into intuitive and rational techniques.
Others 0%
Cultural beliefs – Intuitive 4.44%
Failure mode and effect analysis – Rational 4%
Monte Carlo simulation/analysis – Rational 0.00%
Risk breakdown analysis – Rational 8.89%
Personal feelings and values – Intuitive 9%
Probability matrix – Rational 8.89%
Gut feeling – Intuitive 8.89%
Fish bone – Rational 0%
Brainstorming – Intuitive 13.33%
Worse/best case scenario – Intuitive 13.33%
Personal experiences (heuristics) – Intuitive 11.11%
Fault tree – Rational 4%
Risk register – Intuitive 13.33%
Figure C.2 Risk management techniques known by the participants (all responses).
C.3 Research Proposition 3: Findings 315
Rational,
26.67%
Intuitive,
73.33%
Figure C.3 Risk management techniques known by the participants (categorisation into intuitive
and rational).
Interview Question 9 Which of the following risk management techniques have you
applied on this project? (Please indicate with a tick)
Risk register ……………. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) …………… Risk breakdown analysis …………….
Worse/best case scenario ……………… Monte Carlo simulation/analysis ……………
Brainstorming……………. Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ………… Other (please specify) ……………………………
Findings: The responses against the applied risk management techniques have been
presented in Figure C.4. Figure C.5 categorises the responses into intuitive and rational
techniques.
Others 0%
Cultural beliefs – Intuitive 5.41%
Failure mode and effect analysis – Rational 0%
Monte Carlo simulation/analysis – Rational 0%
Risk breakdown analysis – Rational 10.81%
Personal feelings and values – Intuitive 10.81%
Probability matrix – Rational 8.11%
Gut feeling – Intuitive 8.11%
Fish bone – Rational 0%
Brainstorming – Intuitive 10.81%
Worse/best case scenario – Intuitive 16.22%
Personal experiences (heuristics) – Intuitive 13.51%
Fault tree – Rational 0%
Risk register – Intuitive 16.22%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00% 18.00%
Intuitive,
96.30%
Interview Question 10 On a scale of 1–3, please rank the top three (3) techniques which
you would prefer to use on your future risk management practices? You may select from
the following list – Risk register, Probability matrix, Fault tree, Personal feelings and values,
Personal experiences (heuristics), Risk breakdown analysis, Worse/best case scenario, Monte
Carlo simulation/analysis, Brainstorming, Failure mode and effect analysis, Fish bone, Cul-
tural believes, and Gut feeling (please note: A scale of 1 represents the highest preference, and
a scale of 3 represents the lowest preference).
1st ranked technique .………………………………………………………………………
2nd ranked technique ………………………………………………………………………
3rd ranked technique ………………………………………………………….……………
Findings: The responses against the preferred risk management techniques have been
presented in Figure C.6. Figure C.7 categorises the responses into intuitive and rational
techniques.
Others 0%
Cultural beliefs – Intuitive 0.00%
Failure mode and effect analysis – Rational 0%
Monte Carlo simulation/analysis – Rational 0%
Risk breakdown analysis – Rational 1.89%
Personal feelings and values – Intuitive 0.00%
Probability matrix – Rational 17%
Gut feeling – Intuitive 1.89%
Fish bone – Rational 0%
Brainstorming – Intuitive 11.32%
Worse/best case scenario – Intuitive 5.66%
Personal experiences (heuristics) – Intuitive 16.98%
Fault tree – Rational 2%
Risk register – Intuitive 43.40%
Intuitive,
79.25%
A trend analysis of the responses on the risk management techniques known, applied,
and preferred have been carried out to see if there are any pattern in the responses that could
shed light on the rationale behind the present risk management practices. The evidence has
been presented in Figures C.8 and C.9.
Others 0
0
0
Cultural beliefs – Intuitive 0 2
2
Failure mode and effect analysis – Rational 0
0 2
Monte Carlo simulation/analysis – Rational 0
0
0
Risk breakdown analysis – Rational 1 4
4
Personal feelings and values – Intuitive 0 4
4
Probability matrix – Rational 3 9
4
Gut feeling – Intuitive 1 34
Fish bone – Rational 0
0
0
Brainstorming – Intuitive 4 6
6
Worse/best case scenario – Intuitive 3 6
6
Personal experiences (heuristics) – Intuitive 5 9
5
Fault tree – Rational 012
Risk register – Intuitive 6 23
6
0 5 10 15 20 25
Figure C.8 Trend analysis of the risk management techniques known, applied, and preferred.
The evidence from the trend analysis has revealed high responses for intuitive techniques
across the three assessment variables. The individual response also confirms sole reliance
on intuitive risk management practices for the project manager and contracts manager,
with the remaining differentiated specialist roles exhibiting comparative superior reliance
on the intuitive techniques.
A significant discovery from the trend analysis is the variance between the percentages
of rational techniques known, applied and preferred. Out of the 26.67% ‘rational’ segment
318 Appendix C Case 3 Data Presentation
70.00%
60.00%
50.00%
40.00%
26.67%
30.00%
18.92% 20.75%
20.00%
10.00%
0.00%
Known Applied Preferred
Intuitive 73.33% 81.08% 79.25%
Rational 26.67% 18.92% 20.75%
Figure C.9 Trend analysis of the risk management techniques known, applied, and preferred
(categorisation into intuitive and rational).
of the ‘known’ risk management techniques, only 70.94% are being applied on the case
study project, representing 18.92% of the responses on ‘applied’ risk management tech-
niques. There is however a 77.80% preference to apply the ‘known’ rational techniques
on a future project, representing 20.75% of the responses on ‘preferred’ risk management
techniques. The empirical evidence has therefore revealed that though there appears to be
limited knowledge of rational risk management techniques among the case study partici-
pants, the comparatively higher preference for applying the ‘known’ rational techniques on
future projects suggests the cause of the dominated intuitive practices within the case to be
other extraneous factors beside the disfavour of rational techniques. The evidence of low
level of training and education in project risk management tools and techniques presented
in Figure C.1 could be one of the extraneous factors.
Notwithstanding the above, the comparatively lower preference for future application of
rational risk management techniques implies that the risk management practices of the
research participants within the short term will continue to utilise intuitive thinking and
decision-making.
The empirical evidence on the factors influencing the participants’ preference for the
various risk management techniques have been addressed in Interview Question 11.
Interview Question 11 What are the factors influencing your preference of the above risk
management technique?
1st ranked technique .………………………………………………………………………
2nd ranked technique ………………………………………………………………………
3rd ranked technique ………………………………………………………….……………
Findings: The findings have been presented in Table C.11.
C.3 Research Proposition 3: Findings 319
Case study 3
Theoretical Issue 7
– Theoretical issue 7: Psychometric research findings suggest that most people relate better
to narratives, models, illustrations, and anecdotal simulations better than statistical data
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for
qualitative data presentation formats, over statistical data, and quantitative models.
The related Interview Questions are 12–14 (see Appendix A). The study data has
been examined for evidence on preference for construction risk management data pre-
sentation format. The interview protocol was designed to stimulate intuitive thinking
and decision processing by allowing the participants limited time to respond to the
questions.
Findings: The findings from case study 3 have been summarised in Figures C.10–C.17.
320 Appendix C Case 3 Data Presentation
Interview Question 12 Which of the following forms of risk data presentations do you
know? (Please tick all the applicable)
Narrative ………. Probabilities …………………
Qualitative statements ………… Whole numbers ……………
Graphs …………. Ratios ……………………………
Fractions ………… Scenarios ………
Simulations …………. Other (please specify) ……………
Findings: The responses against the risk management data presentation formats have
been presented in Figure C.10. Figure C.11 also categorises the responses into qualitative
and quantitative presentation formats.
60.71%
Qualitative Quantitative
C.3 Research Proposition 3: Findings 321
Interview Question 13 Which of the following forms of risk data presentations have you
applied on this project? (Please tick all the applicable)
Narrative ………. Probabilities …………………
Qualitative statements ………… Whole numbers ……………
Graphs …………. Ratios ……………………………
Fractions ………… Scenarios ………
Simulations …………. Other (please specify) ……………
Findings: The responses against the risk management data presentation formats have
been presented in Figure C.12. Figure C.13 also categorises the responses into qualitative
and quantitative presentation formats.
70.59%
Qualitative Quantitative
322 Appendix C Case 3 Data Presentation
Interview Question 14 On a scale of 1–3, please rank the top three (3) risk data presentation
formats which you would prefer to use on your future risk management practices? You may
select from the following list – Narratives, Probabilities, Qualitative statements, Whole num-
bers, Graphs, Ratios, Fractions, Scenarios, and Simulations (please note: A scale of 1 represents
the highest preference, and a scale of 3 represents the lowest preference).
Findings: The responses against the risk management data presentation formats have
been presented in Figure C.14. Figure C.15 also categorises the responses into qualitative
and quantitative presentation formats.
77.55%
Qualitative Quantitative
A trend analysis of the responses on the risk management data presentation formats
known, applied, and preferred have been carried out to see if there are any pattern in the
responses that could shed light on the rationale behind the prevailing risk management
practices. The evidence has been presented in Figures C.16 and C.17.
Others
1
Sketches and mark-ups on drawings – Qualitative 1
2
5
Using different colours to rank severity of risk – Qualitative 1
6
Scenarios – Qualitative 2
4
Ratios – Quantitative 1
2
Whole numbers – Qualitative
10
Probabilities – Quantitative 3
3
2
Simulations – Qualitative 1
3
Fractions – Quantitative 1
3
1
Graphs – Quantitative
3
13
Qualitative statement – Qualitative 3
4
11
Narrative – Qualitative 4
4
0 2 4 6 8 10 12 14
Figure C.16 Trend analysis of the risk management data presentation formats known, applied, and
preferred.
324 Appendix C Case 3 Data Presentation
90.00%
77.55%
80.00%
70.59%
70.00% 60.71%
60.00%
50.00%
39.29%
40.00%
29.41%
30.00% 22.45%
20.00%
10.00%
0.00%
Known Applied Preferred
Qualitative 60.71% 70.59% 77.55%
Quantitative 39.29% 29.41% 22.45%
Figure C.17 Trend analysis of the risk management data presentation formats known, applied, and
preferred (categorisation into qualitative and quantitative).
The evidence from the trend analysis has revealed high responses for qualitative data
presentation formats across the three assessment variables, suggesting prolonged existence
for the study participants’ current risk management practice of qualitative data process-
ing. The application and preference of qualitative data formats were elaborated in all the
study participants’ responses. The project manager in particular attributed his attraction
to qualitative data, to the ease of recalling previous experiences to guide spontaneous risk
assessments.
A major concern however is the variance between the proportions of applied systems
of risk management techniques presented under Theoretical Issue 6 (81.08% intuitive ver-
sus 18.92% rational) and the applied systems of risk data formats (70.56% qualitative ver-
sus 29.41% quantitative), which confirms theoretically incompatible practices of intuitive
processing of some quantitative data formats. For instance, the contracts’ manager whose
responses from Theoretical Issue 6 confirmed sole reliance on intuitive techniques has also
admitted quantitative probability data application. The previous theoretical discussion on
the requirement for complementarity between a decision processing vehicle and the applied
data presentation format (Kahneman 2011) suggests the potential for psychological con-
flicts and systematic errors in the contracts manager’s risk analysis output.
Theoretical Issue 8
– Theoretical issue 8: Most people do not trust, understand, or use statistics and probability
estimates in their risk management decision-making (March and Shapira 1987; Akintoye
and Macleod 1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
Theoretical issue to be confirmed if the empirical evidence reveals low level of knowledge
and application of statistics and probability.
C.3 Research Proposition 3: Findings 325
The related Interview Questions are 15–18 (see Appendix A). The study data has been
examined for evidence on the participants’ level of knowledge and understanding of statis-
tics and probability assessment. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions intuitively, without consulting any reference source.
Findings: The findings from case study 3 have been summarised in Table C.12 and
Figures C.18–C.20.
Interview Question 15 Which of the following statistical symbols are you familiar with?
Please indicate by selecting from the following list: summation, probability density func-
tion, probability function, sample mean, population mean, median, variance, standard devi-
ation, sample standard deviation, and don’t know.
Symbol Name
P(A)
S
∑
Case study 3
Interview Question 16–18: Please use the hypothetical risk evaluations represented in the
following table to answer questions 16–18.
A 1/7 30/35
B 97.9% 84%
C 1 : 9743 4:5
326 Appendix C Case 3 Data Presentation
50.00% 50.00%
Interview Question 17 Which of the above risk events is certain not to occur?
Risk event A ………………………………………….
Risk event B ………………………………………….
Risk event C ………………………………………….
All (risk events A, B, and C) …………………………
None of the risk events ………………………………
Don’t know …………………………………………...
Findings: The percentage of correct responses has been presented in Figure C.19.
50.00% 50.00%
Interview Question 18 Which of the above risk events is certain to have the highest impact?
Risk event A ………………………………………….
Risk event B ………………………………………….
Risk event C ………………………………………….
All (risk events A, B, and C) …………………………
None of the risk events ………………………………
Don’t know …………………………………………...
Findings: The percentage of correct responses has been presented in Figure C.20.
83.33%
The empirical evidence has revealed significant low level of understanding of statistical
symbols and probability assessments among the case study participants:
– 2 out of 24 correct identification of statistical symbols which represents 8.33% (see
Table C.12).
– 7 out of 18 correct responses for the probability assessments, which represent 38.89%.
The responses provided by some of the study participants suggested an erroneous belief
of ‘higher probability prediction’ being equal to ‘certainty of an event occurring’, which
may have resulted from a general lack of understanding of the probability concept. Never-
theless, the contracts manager gave correct responses for all the quantitative probability
assessments. At first glance, this may appear contrary to the previous theoretical predic-
tion of potential systematic errors in his intuitive processing of quantitative probability
data, discussed under Theoretical Issue 7. A deeper analytical examination, however sug-
gest the deviation may have resulted from his 16 years extensive industry experience, and
educational background including active participation in continuous professional devel-
opment sessions and seminars on quantitative risk management systems, which over the
years may have become assimilated into his grounded heuristics. The empirical evidence
on the contracts manager’s probability assessment in effect appears consistent with the
existing behaviour science evidence on experts relying on their professional experiences
to achieve accurate intuitive assessment (Kahneman 2011).
328 Appendix C Case 3 Data Presentation
Theoretical Issue 9
– Theoretical issue 9: Most people relate better to qualitative assessments than quantitative
assessments (Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals high accuracy for risk
analysis expressed in whole numbers, qualitative statements, and scenarios compared to
risk analysis expressed in percentages, decimals, and ratios.
The related Interview Questions are 19 and 20 (see Appendix A). The study data has been
examined for evidence on the study participants’ level of knowledge and understanding of
qualitative and quantitative assessments. The interview protocol was designed to stimulate
intuitive thinking and decision processing by allowing the participants limited time to
respond to the questions intuitively without using calculators.
Findings: The findings from case study 3 have been summarised in Figures C.21 and C.22
and Table C.13.
Interview Question 19 Which of the following hypothetical project settings offer the high-
est probability estimate (Indicate with a tick)
A. There is a 75.2% probability of half of the site being contaminated (……)
B. There is a 1 : 321 probability of the site being contaminated (……)
C. There is a 0.381 probability of the site being contaminated (……)
Findings: The percentage of correct responses has been presented in Figure C.21.
66.67%
Interview Question 20 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. The probability of projects over running their initial budget is 3 out of 10 (……)
B. out of every projects overrun their initial budget (……)
C. After reviewing the financial records of the four completed construction projects, only
project A recorded a final cost higher than the initial budget estimate (……)
Findings: The percentage of correct responses has been presented in Figure C.22.
C.3 Research Proposition 3: Findings 329
66.67%
The empirical evidence has revealed higher percentage of accuracy for the qualitative prob-
ability equations compared to the quantitative probability equations:
– Qualitative probability equation: Four out of six correct responses representing
66.67%.
– Quantitative probability equation: Two out of six correct responses representing 33.33%.
Another significant evidence is the composition of the wrong responses. All the four
wrong respondents selected the probability equation expressed in percentages. Further
questioning on the rationale behind the preference for the ‘perc entage equation’ format
revealed the following responses:
a. Percentages are what we are familiar with from our primary and secondary educa-
tion – one response.
b. Percentages are what we use in our everyday working life – three responses.
The above responses suggest that the participants who selected the percentage
equation were simply relying on their availability heuristics to intuitively generate the
answers.
There were two further supplementary questions to solicit empirical evidence on the
participants’ experience with answering the two forms of probability equations. The
findings have been presented in Table C.13. The empirical evidence has revealed
the qualitative equation format to be comparatively straightforward and easy to
understand.
330 Appendix C Case 3 Data Presentation
Table C.13 Participants’ experience with answering the quantitative and qualitative probability
equations.
Theoretical Issue 10
– Theoretical issue 10: Where a potential outcome of a future event evokes a strong affec-
tive memory, the assessor may be insensitive to the probability predictions on the event
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals insensitivity to differ-
ent ranges of hypothetical probability predictions of asbestos contamination risk.
The related Interview Questions are 21i–21iii (see Appendix A). The study data has been
examined for evidence on the participants’ responses to the different range of hypothetical
asbestos contamination risk. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions.
Findings: The findings from case study 3 have been summarised in Table C.14.
Interview Questions 21
Asbestos related illness has been a major cause of construction occupational death. Accord-
ing to the Health and Safety Executive in the United Kingdom, there has been a recent surge
in asbestos related illness and deaths (see the table and pictures below).
1968 153
2009 2321 411
2010 2347 412
2011 2291 429
2012 2535 464
Source: Shutterstock.com
21i Please kindly review the information above and complete the ‘Appropriate risk
response’ column of the ‘Hypothetical project risk evaluation matrix’ below. Please select
your response from any of the following options:
i. Appoint a licenced contractor to carry out further surveys to identify and remove the
asbestos before the commencement of construction works.
ii. Carry on construction works at the non-contaminated area whilst carrying out further
investigations at the remaining areas to identify and remove any asbestos residue.
iii. Supply protective masks to all operatives and carry on with the construction works.
iv. Carry on with the construction works and forget about any asbestos.
21ii\ What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
21iii What will be your risk response if the risk evaluation results were to change to:
(Please select your response from any of the options given under 21i.)
Findings: The participants’ risk responses under the different probability predictions have
been presented in Table C.14.
C.3 Research Proposition 3: Findings 333
Table C.14 The impact of probability predictions on events which evokes strong affective
memories.
Probability of Probability of
Hypothetical occurrence based impact based Appropriate risk Responses/
setting on initial survey on initial survey response percentages
The empirical evidence has revealed significant insensitivity to the different ranges of
hypothetical probability predictions on asbestos contamination. 50% of the participants
consistently responded that they would appoint a licensed contractor to carry out further
asbestos surveys to identify and remove asbestos contaminated materials before they com-
mence site construction works. The different levels of probability risk predictions could
not alter their responses, not even for the very low probability prediction of non-malignant
contamination within a small specified area. The client development manager in particu-
lar indicated that he will always insist on the contractor appointing a specialist firm to fully
address asbestos risk before commencing site construction works, irrespective of the prob-
ability predictions. He was not prepared to compromise on the health and safety of the site
operatives.
The empirical evidence from the direct observations also revealed some level of emo-
tional discomfort for some of the participants during the interrogations, especially the sta-
tistical and probability assessments (Theoretical Issue 8) and the quantitative equations
334 Appendix C Case 3 Data Presentation
(Theoretical Issue 9). Some of the participants even inquired if they could search the statis-
tical symbols on the internet and also process the quantitative equations with calculators.
When this was denied, the frustration became evident in their body language, confirming
the psychological conflict they were facing in attempting intuitive processing of quantitative
data formats.
In summary, the empirical evidence from Theoretical Issues 6–10 has provided support
for research proposition 3. There has been evidence of intuitive dominated construction
risk management practices, and limited risk management training comprising, on the job
informal training, and participation in continuous professional development sessions. The
findings have also revealed massive qualitative data application and preference. Analysis of
the empirical evidence has also revealed intuitive processing of theoretically incompatible
quantitative data formats especially quantitative probability, by the study participants who
have demonstrated low proficiency in statistical symbols and probability judgement. Apart
from the project manager, contracts manager, and quantity surveyor, the remaining differ-
entiated specialist roles demonstrated a general lack of understanding of the probability
concept. The ensuing systematic issues evident from the empirical findings are compar-
atively higher accuracy for the intuitive evaluation of qualitative risk data formats over
quantitative risk data formats and insensitivity to probability predictions of emotive events.
335
Appendix D
Appendix D presents the empirical evidence from case study 4. The findings have been pre-
sented under three subsections corresponding to the three research propositions developed
in Section 6.2. The findings for the propositions have been further organised under separate
headings relating to the underlining theoretical issues. The presentation has adopted the
question and answer approach. The interview questions relating to each theoretical issue
have been described in Appendix A.
Theoretical Issue 1
– Theoretical issue 1: Systems’ differentiation also generates differences in risk perceptions
among the system components – different specialist roles (Walker 2007, 2015; Kahneman
2011; Slovic et al. 2010).
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
336 Appendix D Case 4 Data Presentation
Interview Question 22 Drawing from your professional experiences, what will you say are
the factors likely to affect project success (risk events) at the pre-construction phase (before
the start of actual construction on site)?
Findings: The responses have been presented in rankings corresponding to the order in
which the participants provided the risk events. A score of 5 has been assigned to the first
generated risk event, 3 for the second generated risk event, and 1 for the third generated risk
event. The risk events have also been categorised using the table of construction project risk
events (Table 7.3) developed from the cumulative responses of interview questions 22–24,
and structured along the lines of a similar previous classification by Perry and Hayes (1985).
The empirical findings have been summarised in Table D.1.
The differentiated builder specialist subgroup comprising, the technical manager, project
manager, and contracts manager generated risk perceptions in the following order of rank-
ing: ground condition related (first), design and quality related (second), legal and statutory
approval related (third). The designers comprising the engineer and architect also gener-
ated risk perceptions in the following order of ranking: legal and statutory approval and
construction related (first), financial and commercial and logistics and commercial related
(second), and political and design and quality related (third). The quantity surveyor’s gener-
ated risk perceptions on the other hand were in the following order of ranking: ground con-
dition related (first), design and quality related (second), and construction related (third).
The client generated risk perceptions were also of the following ranking: ground condition
related (first) and design and quality related (second).
The empirical findings have subsequently revealed evidence of differentiation within the
risk perceptions at the pre-construction phase. Apart from ground condition related risk
events which appeared common in most of the responses, the remaining generated risk
perceptions exhibited reasonable correlation with the study participants’ specialist back-
ground. The universality of ‘ground condition related’ risk perceptions may have resulted
from the dominant influence of an associated construction industry affective heuristics
component (Finucane et al. 2003; Slovic et al. 2010) within the study participants’ risk
identification transformational systems (Walker 2007, 2015), leading to the generation of
associated ‘ground condition related’ risk perceptions (Benthin et al. 1993).
D.1 Research Proposition 1: Findings 337
Interview Question 23 Drawing from your professional experiences, what will you say are
the factors likely to affect project success (risk events) at the construction phase (during
actual construction work on site)?
Findings: The responses have been presented in Table D.2 using the same format as the
previous question Interview Question 22.
The empirical evidence has once again revealed differentiation within the risk percep-
tions at the construction phase, with a pattern consistent with the differences in special-
ist backgrounds. The differentiated builder specialist subgroup comprising, the technical
manager and project manager generated risk perceptions in the following order of rank-
ing: financial and commercial related (first), construction and legal and statutory approval
related (second), and design and quality related (third). The designers comprising the engi-
neer and architect also generated risk perceptions in the following order of ranking: design
and quality related (first) and construction related (second). The quantity surveyor’s gen-
erated risk perceptions on the other hand were in the following order of ranking: financial
and commercial related (first) and design and quality related (second). The client gener-
ated risk perceptions in the following order of ranking: design and quality related (first)
and construction related (second).
Theoretical Issue 2
– Theoretical issue 2: The differences in objectives of differentiated subsystems
(Carmichael 2006; Loosemore et al. 2006) suggest that the impact of environmen-
tal forces will result in differences in affective heuristics, and therefore differences in
perceptions within the system components – different specialist roles.
Theoretical issue to be confirmed if the empirical evidence reveals similarities in the
respondents’ risk perceptions on hypothetical project settings and their specialist back-
ground.
The related interview question is numbered Interview Question 24 (see Appendix A).
The rationale behind this question was to solicit the participants’ risk perceptions on two
hypothetical construction project settings.
D.1 Research Proposition 1: Findings 339
Interview Question 24 The pictures below depict hypothetical sites (edged in bold front)
being considered for residential development. The project is at the inception stage with the
following key project objectives:
– Design: 50 units to meet minimum building regulation requirements, lifetime homes,
secured by design, minimum 40% carbon reduction.
– Budget: Construction cost not to exceed £15 million.
– Programme: To be completed within two years.
Hypothetical site A: Brown field development within high density urban area
Please list the factors that are likely to affect the success of the projects (risk events) during
the pre-construction and actual construction phases.
Findings: The responses outline the participants’ risk perceptions on the hypothetical
projects. The risk events have been categorised using the table of construction project risk
events (Table 7.3). The findings have been summarised in Tables D.3–D.5.
The data from the two hypothetical project settings have been further analysed for evi-
dence of consistency in responses and also to filter out the other extraneous risk formation
variables which are unrelated to specialist role affective heuristics.
The empirical findings have revealed evidence of differentiation within the generated
risk perceptions from the two hypothetical project settings, with the study participants suc-
cessfully identifying risk events from related specialist categories. The consistency analysis
has revealed comparatively broader categorisation in most of the participants’ risk identifi-
cation. This may have resulted from their long years of involvement in the construction
industry as evident in Table 7.23 (see Chapter 7), and the consequential propensity for
multi-specialist exposures. The project manager’s initial involvement in the construction
industry was in quantity surveying, whilst the client development manager also previously
worked as a roofer. Their extensive construction industry experience may have resulted in
the availability of wide-ranging grounded heuristics, influencing the generation of compa-
rably broader categorisation of risk events.
The findings from the consistency analysis have been presented below:
– Ground condition related risk events were perceived by the quantity surveyor, technical
manager, and contracts manager.
– Design and quality related risk events were perceived by the quantity surveyor, technical
manager, architect, and client development manager.
– Logistics and communication related risk events were perceived by the client develop-
ment manager, project manager, and engineer.
– Financial and commercial related risk events were perceived by the quantity surveyor,
architect, and contracts manager.
– Legal and statutory approval related risk events were perceived by the quantity surveyor,
technical manager, architect, client development manager, and contracts manager.
– Construction related risk events on the other hand were perceived by the architect and
engineer.
– The client development manager was the only research participant to perceive opera-
tional related risk events.
Evaluation of the evidence from the consistency analysis from the perspective of the dif-
ferent specialist subgroups has also revealed the following categorisations:
– The builders comprising the technical manager, the project manager, and the contracts
manager have exhibited varied risk perceptions with the dominant consistent events
being ground condition, and legal and statutory approval related.
– The designers comprising the architect and the engineer also displayed risk perceptions
relating to design and quality, logistics and communication, financial and commercial,
legal and statutory approval, and construction events. Construction related risk events
were consistent in all the responses.
342 Appendix D Case 4 Data Presentation
Case study 4
Cumulative Cumulative
Risk perception TM-4 PM-4 CM-4 for builders A-4 E-4 for designers QS-4 CDM-4
Force majeure related
Case study 4
Cumulative Cumulative
Risk perception TM-4 PM-4 CM-4 for builders A-4 E-4 for designers QS-4 CDM-4
Legal and statutory approval
related
Achieving planning Yes Yes Yes
permission and discharging
conditions
Land title and boundaries Yes
Partywall issues resulting Yes Yes Yes
from adjoining properties
Statutory approvals Yes Yes Yes
Political related
Community engagement Yes Yes
Construction related
Programming of construction Yes Yes Yes Yes
works to achieve completion
date
Safety on site Yes Yes
Operational related
Insufficient social Yes
facilities – schools, GP, etc.
Case study 4
Cumulative Cumulative
Risk perception TM-4 PM-4 CM-4 for builders A-4 E-4 for designers QS-4 CDM-4
Force majeure related
(Continued)
344 Appendix D Case 4 Data Presentation
Case study 4
Cumulative Cumulative
Risk perception TM-4 PM-4 CM-4 for builders A-4 E-4 for designers QS-4 CDM-4
Design and quality related
Availability and feasibility of Yes Yes Yes Yes Yes Yes Yes Yes
services connections
Coordinating design with Yes Yes
architects and other trades
Political related
Construction related
Programming construction Yes Yes
works to achieve project
completion date
Safety on site Yes Yes
Form of construction Yes
Operational related
Insufficient social Yes
infrastructure – e.g. school,
GP, etc.
Table D.5 Consistency analysis of the risk perceptions from hypothetical projects A and B (Case 4).
Case Study 4
Questions & Responses 24A 24B Consistency 24A 24B Consistency 24A 24B Consistency 24A 24B Consitency 24A 24B Consistency 24A 24B Consistency 24A 24B Consitency 24A 24B Consistency 24A 24B Consistency
Ground Condition
Related
Availability and Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
feasibility of services
connections
(Continued)
Table D.5 (Continued)
Case Study 4
Questions & Responses 24A 24B Consistency 24A 24B Consistency 24A 24B Consistency 24A 24B Consitency 24A 24B Consistency 24A 24B Consistency 24A 24B Consitency 24A 24B Consistency 24A 24B Consistency
Logistics &
Communication
Related
Access and egress into Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
the site
Site setup and Yes Yes Yes Yes Yes Yes Yes
organisation
Financial &
Commercial Related
Completing within Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
budget
Subcontractor Yes
procurement
Achieving planning Yes Yes Yes Yes Yes Yes Yes Yes Yes
permission &
discharging conditions
Statutory approvals Yes Yes Yes Yes Yes Yes Yes Yes
Political Related
Construction Related
Operational Related
Yes
348 Appendix D Case 4 Data Presentation
– The quantity surveyor on the other hand has exhibited risk perceptions relating to
financial and commercial, ground condition, design and quality, and legal and statutory
approval events.
– The client has also displayed risk perceptions associated with design and quality, logistics
and communication, legal and statutory, and operational related events.
In summary, the empirical evidence from Theoretical Issues 1 and 2 have supported
research proposition 1. There has been evidence of differentiation within the study partici-
pants’ responses on the typical project risk events. Apart from ground condition related risk
events which appeared common in the responses on the typical pre-construction project
risk events, the remaining answers exhibited reasonable correlation with the study par-
ticipants’ specialist backgrounds. The empirical evidence on the risk identification under
different hypothetical project settings also generated a broad range of risk perceptions, with
a pattern reflecting the differences in specialist background. The realisation of consistent
patterns of differentiation at the different project delivery phases and under different project
settings have inferably provided reasonable validation for the study’s novel theoretical pre-
diction of risk perception categorisation within the construction project delivery system
reflecting the differences in specialist role affective heuristics.
The second research proposition advances that intuitive risk management practices under
the representative and availability heuristics, which exclude some of the project subgroup-
ings, run the risk of failing to adequately identify potential risk events. The underlining
theoretical issues are as follows:
– 3. Perceptions are formed from the representative and availability heuristics (Kahneman
2011; Slovic et al. 2010).
– 4. The inherent subjectivity of perceptions makes it manipulative and less rational. Addi-
tional information could change risk perception even where the base situation has not
changed – best-case scenarios which lead to gaps between pre-commencement estimates
and actual values at completion stage (Flyvbjerg et al. 2003; Alhakami and Slovic 1994;
Slovic and Peters 2006).
– 5. An uncertainty becomes a risk event to a subject if its occurrence can potentially affect
the objective(s) of the subject (Loosemore et al. 2006), suggesting that where there is no
possibility of impact, the subject may fail to recognise the uncertainty as a risk event.
The presentation of findings has been organised under three headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants supported the study theoretical predictions.
D.2 Research Proposition 2: Findings 349
Theoretical Issue 3
– Theoretical issue 3: Perceptions are formed from the representative and availability
heuristics (Kahneman 2011; Slovic et al. 2010).
Theoretical issue to be confirmed if the participants’ risk identification on the hypotheti-
cal project settings correspond to their availability heuristics (evident from their responses
on the common construction project risk events).
The related interview questions are 22–24 (see Appendix A). The study data have been
examined for evidence of the influence of grounded heuristics in the risk identification
processes, using hypothetical construction project settings. The interview protocol was
designed to first collect data on the participants’ grounded heuristics before asking them
to identify risk events from hypothetical project settings. The rationale was to ascertain
whether the risk identification on the hypothetical project settings was done intuitively
through reliance on heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
grounded heuristics, against their risk identification on hypothetical project settings. The
risk events have also been categorised using the table of construction project risk events
(Table 7.3). The findings have been summarised in Table D.6.
Evidence from the empirical findings suggests substantial risk identification based on the
participants’ grounded heuristics:
– QS-4: Three out of the five risk identification from hypothetical project A have corre-
sponding grounded heuristics. Three out of the four risk identification from hypothetical
project B have corresponding grounded heuristics.
– TM-4: Three out of the three risk identification from hypothetical project A have corre-
sponding grounded heuristics. Four out of the four risk identification from hypothetical
project B have corresponding grounded heuristics.
– A-4: Four out of the five risk identification from hypothetical project A have correspond-
ing grounded heuristics. Three out of the six risk identification from hypothetical project
B have corresponding grounded heuristics.
– CDM-4: One out of the four risk identification from hypothetical project A has corre-
sponding grounded heuristics. One out of the four risk identification from hypothetical
project B has corresponding grounded heuristics.
– CM-4: Three out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Four out of the four risk identification from hypothetical
project B have corresponding grounded heuristics.
– PM-4: Three out of the four risk identification from hypothetical project A have corre-
sponding grounded heuristics. Two out of the three risk identification from hypothetical
project B have corresponding grounded heuristics.
– E-4: Three out of the three risk identification from hypothetical project A have corre-
sponding grounded heuristics. Two out of the two risk identification from hypothetical
project B have corresponding grounded heuristics.
Table D.6 Analysis of perception being formed from availability and representative heuristics (Case 4).
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Questions &
Responses 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 22B 22 & 23 24A 24B 22 & 23 24A 24B
Force Majeure
Related
Weather condition Yes
Ground Condition
Related
Contamination of site Yes Yes Yes
Excavation of site Yes
Previous landuse to Yes Yes Yes Yes
establish ground
conditions
Site constraints, eg Yes Yes
congested site
Ecological issues Yes Yes Yes Yes
Ground investigations Yes
Buried services Yes
Yes Yes Yes Yes No Yes Yes Yes
Design & Quality
Related
Getting the design Yes Yes
right at the concept
stage
Availability and Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
feasibility of service
connections
Impact of existing Yes
service layouts on
design
Poor quality control Yes
Managing client Yes Yes
design expectations
Co ordination of Yes Yes
design team
Availability of Yes Yes
specialist contractors
to build to design
Right of Light Yes
Client design changes Yes
Sustainability- Yes
renewable system
Proximity to transport Yes Yes
network- road; rail
lines (obtaining
relevant approvals)
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Logistics &
Communication
Related
Site access Yes Yes Yes Yes Yes Yes
Communication and Yes
information flow
Site setup and Yes Yes Yes
organisation of
construction plants
Site setup, welfare Yes
and deliveries
Delivery of material Yes
to the site
(Continued)
TableD.6 (Continued)
Case study 4
QS-4 TM-4 A-4 CDM-4 CM-4 PM-4 E-4
Questions &
Responses 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 24B 22 & 23 24A 22B 22 & 23 24A 24B 22 & 23 24A 24B
Availability of Yes Yes
specialist labour
Site parking facilities Yes
No No No No No No No Yes Yes
Financial &
Commercial Related
Procurement of Yes
labour and material
Completing within Yes Yes Yes Yes
budget
Availability of funds Yes Yes
Supply chains Yes
management
Subcontract Yes Yes Yes Yes
procurement
Incorrect project Yes
pricing at pre-
contract stage
No No
3 of 5 3 of 4 3 of 3 4 of 4 4 of 5 3 of 6 1 of 4 1 of 4 3 of 4 4 of 4 3 of 4 2 of 3 3 of 3 2 of 2
354 Appendix D Case 4 Data Presentation
Theoretical Issue 4
– Theoretical issue 4: The inherent subjectivity of perception makes it manipulative
and less rational. Additional information could change risk perception even where
the base situation has not changed – best-case scenarios which lead to gaps between
pre-commencement estimates and actual values at completion stage (Flyvbjerg et al.
2003; Alhakami and Slovic 1994; Slovic and Peters 2006).
Theoretical issue to be confirmed if exposure to additional information (a schedule of
construction risk sources) changes the participants’ risk perception/analysis.
The related interview questions are Interview Question 24 and 25 (see Appendix A). The
study data has been examined to see if exposure to additional information (a schedule of
construction risk sources) can influence the risk identification process. The interview pro-
tocol was designed to first solicit the participants’ risk perceptions on hypothetical project
settings using their heuristics. The participants were later given a schedule of construction
risk sources to study and re-identify risk events from the same hypothetical project settings.
The rational was to ascertain whether the introduction of additional information will alter
their previous risk evaluation. The risk identifications after the participants were provided
with additional information have been termed exposed heuristics, denoting that they have
been generated based on exposure to additional information which has become assimilated
within the pool of heuristics applied in their intuitive evaluations.
Findings: The responses have been presented in matching patterns of the respondents’
initial risk identification on the hypothetical project settings, against their subsequent risk
identification on the same hypothetical project settings following exposure to additional
assessment information. The risk events have also been categorised using the table of con-
struction project risk events (Table 7.3). The findings have been summarised in Table D.7.
The empirical evidence has revealed changes in the initial risk identification profiles
after the participants were exposed to additional assessment information, thereby confirm-
ing the susceptibility of perceptions to manipulation. The changes occurred even though
the project settings have not been altered. The changes in risk identification profiles were
observed in the responses from all the participants apart from the client development man-
ager who declined to re-assess the hypothetical project settings after receiving the additional
assessment information and simply responded that his answers to interview question 25 are
the same as those offered in the initial risk identification.
– QS-4: One out of the five risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. Two out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– TM-4: Two out of the four risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Three out of the four risk assessment categories on
hypothetical project B were different from the initial assessment.
– A-4: One out of the five risk assessment categories on hypothetical project A was different
from the initial assessment. Two out of the five risk assessment categories on hypothetical
project B were different from the initial assessment.
– CDM-4: Zero out of the four risk assessment categories on hypothetical project A was
different from the initial assessment. Zero out of the four risk assessment categories on
hypothetical project B was different from the initial assessment.
Table D.7 Analysis of the influence of additional information on perception formation (Case 4).
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 24A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Force Majeure Related
Fire Yes
Earthquake Yes Yes
Structural collapse Yes Yes
Flood destruction Yes Yes
Yes Yes Yes Yes Yes Yes Yes
Ground Condition
Related
Contamination of site Yes Yes Yes Yes Yes Yes
Excavation of site Yes
Polution from adjoining Yes
sites
Ecological issues Yes Yes Yes Yes Yes Yes Yes
Ground investigations Yes
Existing structures and Yes Yes
foundations
Buried services Yes
No No No No No Yes No Yes Yes
(Continued)
Table D.7 (Continued)
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 24A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Design & Quality
Related
Availability and Yes Yes Yes Yes Yes Yes Yes Yes Yes
feasibility of service
connections
Co ordination of design Yes
team
Availability of specialist Yes Yes
contractors to build to
design
Right of Light Yes
Sustainability- Yes
renewable system
Proximity to transport Yes Yes
network- road; rail lines
(obtaining relevant
approvals)
Accuracy and Yes Yes
appriopriateness of
design specifications
Design constraints and Yes Yes Yes Yes
opportunities arising
from surveys
Site investigations Yes Yes
No No No Yes No Yes No No Yes No
Logistics &
Communication
Related
Site access Yes Yes Yes Yes Yes Yes Yes Yes Yes
Planning and Yes
programming of project
delivery
Site setup and Yes Yes
organisation of
construction plants
Site setup, welfare and Yes
deliveries
Delivery of material to Yes Yes Yes Yes
the site
Availability of specialist Yes Yes Yes Yes Yes
labour
Site parking facilities Yes
No Yes No No No No No No No
Financial &
Commercial Related
Completing within Yes Yes Yes Yes
budget
Availability of funds Yes Yes Yes Yes Yes
Supply chains Yes
management
Subcontract Yes
procurement
Exchange rate Yes Yes
fluctuations
Fluctuations in market Yes Yes Yes
demand
Yes Yes No No No No
(Continued)
Table D.7 (Continued)
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Questions & Responses 24A 25A 24B 25B 24A 25A 24B 25B 24A 24A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B 24A 25A 24B 25B
Legal and Statutory
Approval Related
Achieving planning Yes Yes Yes Yes Yes Yes Yes Yes
permissions &
discharge of conditions
Third party Yes Yes Yes Yes Yes
issues- partywall, etc
Statutory approvals Yes Yes Yes Yes Yes Yes Yes Yes Yes
including building
regulations
Land titles and Yes Yes
boundaries
Differences between Yes
regional and natiional
laws & regulations
No No No No No No No No Yes No Yes
Political Related
Community Yes
engagement
Differences in Yes
stakeholder
expectations
Changes in law and Yes Yes
regulations
Yes Yes Yes
Construction Related
Safety Yes Yes
Achieving project Yes Yes Yes
completion within
programme
Construction technique Yes
Availability and quality Yes Yes Yes
of management and
supervision
Suitability of Yes Yes
construction method
Yes Yes Yes No No
Operational Related
Insufficient social Yes Yes Yes Yes
facilities- schools,
– CM-4: One out of the four risk assessment categories on hypothetical project A was dif-
ferent from the initial assessment. One out of the four risk assessment categories on
hypothetical project B was different from the initial assessment.
– PM-4: Three out of the five risk assessment categories on hypothetical project A were
different from the initial assessment. Four out of the five risk assessment categories on
hypothetical project B were different from the initial assessment.
– E-4: Two out of the three risk assessment categories on hypothetical project A were dif-
ferent from the initial assessment. Three out of the four risk assessment categories on
hypothetical project B were different from the initial assessment.
Theoretical Issue 5
– Theoretical issue 5: An uncertainty becomes a risk event to a subject if its occurrence can
potentially affect the objective(s) of the subject (Loosemore et al. 2006), suggesting that
where there is no possibility of impact, the subject may fail to recognise the uncertainty
as a risk event.
Theoretical issue is to be confirmed if the participants fail to identify risk events outside
their specialist background (associated with lack of related grounded heuristics).
The related interview question is Interview Question 24 (see Appendix A). The study
data have been examined to see if the participants’ failure in identifying certain risk events
can be attributed to their lack of related grounded heuristics outside their specialist back-
ground. The interview protocol was designed to solicit the participants’ risk perceptions on
hypothetical project settings using their grounded heuristics.
Findings: The responses have been presented in matching patterns of the respondents’
risk identification on the hypothetical project settings, against the project cumulative risk
identification profiles, to reveal the items missed by each participant. The risk events have
also been categorised using the table of construction project risk events (Table 7.3). The
findings have been summarised in Tables D.8–D.10.
The data from the two hypothetical project settings has been further analysed for consis-
tency in the participants’ missed risk events.
The empirical evidence and consistency examination have revealed that exclusion of a
specialist role from the risk identification exercise would have resulted in some of the risk
events being missed. The individual responses have revealed some risk events being per-
ceived by singular specialist roles. As an illustration, the client development manager was
the only specialist role to have perceived operational related risk events for the two hypo-
thetical project settings. The quantity surveyor also was the only specialist role to have
perceived procurement of materials to be a risk event to the hypothetical project setting A.
This suggest that exclusion of the client development manager and the quantity surveyor
from the risk identification exercise may have resulted in the missing of the associated risk
events with consequential impact on the ensuing risk management responses and project
performance, assuming the hypothetical projects and risk events to be real.
Empirical evidence from the direct observations also revealed that the research partic-
ipants were generally calm and relaxed during the interrogations relating to Theoretical
Issues 3–5. Their responses were spontaneous and effortless, suggesting intuitive decision
processing (Kahneman 2011).
Table D.8 Effect of limited heuristics (excluding some project team members) on project risk identification – Hypothetical project A.
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Risk Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24A events events events events events events events events events events events events events events events
Force Majeure
Related
Ground
Condition
Related
Demolition & Yes Yes
excavation
Site Yes Yes
contaminations
Buried services Yes Yes
Ground Yes Yes
investigations
Missed Missed Missed Missed
Design & Quality
Related
Availability and Yes Yes Yes
feasibility of
services
connections
Rights of light Yes Yes
(Continued)
Table D.8 (Continued)
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Risk Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24A events events events events events events events events events events events events events events events
Sustainability- Yes Yes
renewable system
Proximity to Yes Yes Yes
transport
network- road;
rail lines-
obtaining
relevant
approvals
Availability of Yes Yes
specialist
contractors to
build to design
Missed Missed
Logistics &
Communication
Related
Access and egress Yes Yes Yes Yes
into the site
Site setup and Yes Yes Yes
organisation
Access for Yes Yes
delivery vehicles
Availability of Yes Yes
specilist labour
Site parking Yes Yes
facilities
Missed Missed
Financial &
Commercial
Related
Completing Yes Yes Yes
within budget
Availability of Yes Yes
funding
Subcontractor Yes Yes
procurement
Missed Missed Missed
Legal & Statutory
Approval Related
Achieving Yes Yes Yes
planning
permission &
discharging
conditions
Land title and Yes Yes
boundaries
Partywall issues Yes Yes Yes
resulting from
adjoining
properties
Statutory Yes Yes Yes
approvals
Missed
Political Related
Community Yes Yes
engagement
Missed Missed Missed Missed Missed Missed
(Continued)
Table D.8 (Continued)
Case study 4
QS-4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Risk Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24A events events events events events events events events events events events events events events events
Construction
Related
Programming of Yes Yes Yes
construction
works to achieve
completion date
Safety on site Yes Yes
Missed Missed Missed Missed
Operational
Related
Insufficient social Yes Yes
facilities- schools,
Case study 4
QS-4 TM-4 A-4 CDM-4 CM-4 PM-4 E-4
Risk Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24B events events events events events events events events events events events events events events events
Force Majeure
Related
Ground
Condition
Related
Ground Yes Yes Yes
contamination
Ecological Yes Yes Yes Yes Yes
Missed Missed
Design & Qulity
Related
Availability and Yes Yes Yes Yes Yes Yes Yes
feasibility of
services
connections
Coordinating Yes Yes
design with
architects and
other trades
Missed
(Continued)
Table D.9 (Continued)
Case study 4
QS-4 TM-4 A-4 CDM-4 CM-4 PM-4 E-4
Risk Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24B events events events events events events events events events events events events events events events
Logistics &
Communication
Related
Access to the site Yes Yes Yes Yes
Availability of Yes Yes
skilled labour
Site set-up/ Yes Yes
establishment
Missed Missed Missed
Financial &
Commercial
Related
Completing Yes Yes Yes
within budget
Supply Yes Yes
chains-
procurement
of materials
Missed Missed Missed Missed
Legal & Statutory
Approval Related
Achieving Yes Yes Yes Yes
planning
permission &
discharging
conditions
Land title/ Yes Yes
ownership
Statutory approvals Yes Yes Yes
Missed Missed
Political Related
Construction
Related
Programming Yes Yes
construction
works to achieve
project
completion date
Safety on site Yes Yes
Form of Yes Yes
construction
Missed Missed Missed Missed
(Continued)
Table D.9 (Continued)
Case study 4
QS- 4 TM- 4 A- 4 CDM- 4 CM- 4 PM- 4 E- 4
Risk Project Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed Individual Missed
Perceptions risk risk risk risk risk risk risk risk risk risk risk risk risk risk risk
of 24B events events events events events events events events events events events events events events events
Operational
Related
Insufficient social Yes Yes
infrastructure- eg
school, GP, etc
Missed Missed Missed Missed Missed Missed
Source: Developed for this research study
Key
Identified risk
events.
Missed risk events–
project level.
D.3 Research Proposition 3: Findings 369
GCR, ground condition related; DQR, design and quality related; LCR, logistics and communication
related; FCR, financial and communication related; LSAR, legal and statutory approval related; PR,
political related; CR, construction and site related; OR, operational related.
In summary, the empirical evidence from Theoretical Issues 3–5 have provided support
for research proposition 2. The study participants’ risk assessment on hypothetical project
settings have produced findings correlating with their grounded heuristics, which confirms
intuitive decision processing inspired by the representative and availability heuristics. All
the study participants, except the client development manager who declined to re-appraisal
the hypothetical project settings, changed their initial risk assessment profiles after being
exposed to additional information in the form of a schedule of construction risk sources. The
variations occurred even though the project settings remained un-altered, which validates
the susceptibility of perceptions to manipulation. A review of the study participants’ indi-
vidual risk assessments against the cumulative case risk identification profile have revealed
missed risk events outside their respective specialist backgrounds, which inferably provides
support for the research proposition of, restricted risk management practices impairing
robust risk identification and decision making, especially the risk categories associated with
the specialist background of the excluded roles.
– 7. Psychometric research findings suggest that most people relate better to narratives,
models, illustrations, and anecdotal simulations better than statistical data (Slovic et al.
2010).
– 8. Most people do not trust, understand, or use statistics and probability estimates in their
risk management decision-making (March and Shapira 1987; Akintoye and Macleod
1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
– 9. Most people relate better to qualitative assessments than quantitative assessments
(Slovic et al. 2010).
– 10. Where a potential outcome of a future event evokes a strong affective memory, the
assessor may be insensitive to the probability predictions on the event (Slovic et al. 2010).
The presentation of findings has been organised under five headings corresponding to
the above theoretical issues. The findings have been analysed to test the extent to which the
empirical evidence from the interviews and direct observations of the six research partici-
pants have supported the study’s theoretical predictions.
Theoretical Issue 6
– Theoretical issue 6: Research findings have identified construction risk management to
be dominated by people without formal training in risk management systems using intu-
ition and heuristics (Akintoye and Macleod 1997; Lyons and Skitmore 2004; Kululanga
and Kuotcha 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for intuitive
risk management approaches over rational tools and techniques.
The related interview questions are 7–11 (see Appendix A). The study data have been
examined for evidence on the level of knowledge, application, and preference for rational
risk management techniques within the construction risk management decision-making
subsystems. The interview protocol was designed to stimulate intuitive thinking and deci-
sion processing by allowing the participants limited time to respond to the questions.
Findings: The findings from case study 4 have been summarised below in Figures D.1–D.9
and Table D.12.
Interview Question 7 Please indicate your level of education/training in project risk man-
agement tools and practices?
a. No formal education or training in project risk management (… . . . .)
b. On the job informal experience (… . . . .)
c. Continuing professional developments (CPDs) and seminars on project risk manage-
ment (… . . . .)
d. Short duration formal course (less than six months) (… . . . .)
e. University diploma/degree in Project Risk Management – e.g. BSc Construction Risk
Management, MSc Risk Management (… . . . .)
Please specify title of degree/diploma…………………………………………………………
Findings: The responses have been presented in Figure D.1.
D.3 Research Proposition 3: Findings 371
University diploma/degree 0%
Interview Question 8 Which of the following risk management techniques have you heard
of? (Please indicate with a tick)
Risk register ………… . . . .. Probability matrix ……………
Fault tree ...………… Personal feelings and values……………
Personal experiences (heuristics) …………… Risk breakdown analysis ………… . . . ..
Worse/best case scenario ……………… Monte Carlo simulation/analysis ………… . . . .
Brainstorming………… . . . .. Failure mode and effect analysis ……………
Fish bone …………… Cultural beliefs ……………
Gut feeling ……… . . . .. Other (please specify) …………………………
Findings: The responses against the risk management techniques have been presented in
Figure D.2. Figure D.3 also categorises the responses into intuitive and rational techniques.
Interview Question 9 Which of the following risk management techniques have you
applied on this project? (Please indicate with a tick)
Others 0%
Cultural beliefs – Intuitive 2.38%
Failure mode and effect analysis – Rational 2%
Monte Carlo simulation/ analysis – Rational 0.00%
Risk breakdown analysis – Rational 4.76%
Personal feelings and values – Intuitive 10%
Probability matrix – Rational 7.14%
Gut feeling – Intuitive 14.29%
Fish bone – Rational 0%
Brainstorming – Intuitive 14.29%
Worse/best case scenario – Intuitive 14.29%
Personal experiences (heuristics) – Intuitive 14.29%
Fault tree – Rational 0%
Risk register – Intuitive 16.67%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00%12.00%14.00%16.00%18.00%
Figure D.2 Risk management techniques known by the participants (all responses).
Intuitive,
85.71%
Findings: The responses against the applied risk management techniques have been
presented in Figure D.4. Figure D.5 categorises the responses into intuitive and rational
techniques.
Interview Question 10 On a scale of 1–3, please rank the top three (3) techniques which
you would prefer to use on your future risk management practices? You may select from the
following list – Risk register, Probability matrix, Fault tree, Personal feelings and values, Per-
sonal experiences (heuristics), Risk breakdown analysis, Worse/best case scenario, Monte Carlo
simulation/analysis, Brainstorming, Failure mode and effect analysis, Fish bone, Cultural
D.3 Research Proposition 3: Findings 373
Others 0%
Cultural beliefs – Intuitive 3.13%
Failure mode and effect analysis – Rational 3%
Monte Carlo simulation/analysis – Rational 0%
Risk breakdown analysis – Rational 6.25%
Personal feelings and values – Intuitive 9.38%
Probability matrix – Rational 3.13%
Gut feeling – Intuitive 15.63%
Fish bone – Rational 0%
Brainstorming – Intuitive 15.63%
Worse/best case scenario – Intuitive 12.50%
Personal experiences (heuristics) – Intuitive 12.50%
Fault tree – Rational 0%
Risk register – Intuitive 18.75%
Intuitive,
87.50%
believes, and Gut feeling (please note: A scale of 1 represents the highest preference, and a scale
of 3 represents the lowest preference).
1st ranked technique ………………………………………………………………………….
2nd ranked technique …………………………………………………………………………
3rd ranked technique ………………………………………………………………………….
Findings: The responses against the preferred risk management techniques have been
presented in Figure D.6. Figure D.7 categorises the responses into intuitive and rational
techniques.
374 Appendix D Case 4 Data Presentation
Others 0%
Cultural beliefs – Intuitive 0.00%
Failure mode and effect analysis – Rational 2%
Monte Carlo simulation/analysis – Rational 0%
Risk breakdown analysis – Rational 4.76%
Personal feelings and values – Intuitive 1.59%
Probability matrix – Rational 0%
Gut feeling – Intuitive 6.35%
Fish bone – Rational 0%
Brainstorming – Intuitive 20.63%
Worse/best case scenario – Intuitive 1.59%
Personal experiences (heuristics) – Intuitive 19.05%
Fault tree – Rational 0%
Risk register – Intuitive 44.44%
Intuitive,
93.65%
A trend analysis of the responses on the risk management techniques known, applied,
and preferred have been carried out to see if there are any pattern in the responses that could
shed light on the rationale behind the existing risk management practices. The evidence has
been presented below in Figures D.8 and D.9.
The evidence from the trend analysis has revealed high responses for intuitive techniques
across the three assessment variables. All the study participants except the technical man-
ager and project engineer confirmed sole reliance on intuitive risk management techniques.
The quantity surveyor confessed that notwithstanding the mathematical orientation of his
specialist role, his risk management practices are mostly guided by personal experiences
and gut feelings.
D.3 Research Proposition 3: Findings 375
Others 0
0
0
01
Cultural beliefs – Intuitive 1
Failure mode and effect analysis – Rational 1
1
1
Monte Carlo simulation/analysis – Rational 0
0
0
Risk breakdown analysis – Rational 23
2
Personal feelings and values – Intuitive 1 3
4
Probability matrix – Rational 01
3
Gut feeling – Intuitive 45
6
Fish bone – Rational 0
0
0
Brainstorming – Intuitive 56 13
Worse/best case scenario – Intuitive 1 4
6
Personal experiences (heuristics) – Intuitive 4 6 12
Fault tree – Rational 0
0
0
Risk register – Intuitive 67 28
0 5 10 15 20 25 30
Figure D.8 Trend analysis of the risk management techniques known; applied and preferred.
100.00% 93.65%
85.71% 87.50%
90.00%
80.00%
70.00%
60.00%
50.00%
40.00%
30.00%
20.00% 14.29% 12.50%
6.35%
10.00%
0.00%
Known Applied Preferred
Intuitive 85.71% 87.50% 93.65%
Rational 14.29% 12.50% 6.35%
Figure D.9 Trend analysis of the risk management techniques known; applied and preferred
(categorisation into intuitive and rational).
In summary, the empirical evidence from Theoretical Issue 6 have revealed limited
knowledge and application of rational risk management techniques among the case study
participants, which may be due to the low level of training and education in project risk
management tools and techniques, as presented in Figure D.1. The empirical evidence
therefore suggests that the risk management practices of the research participants within
the short term will continue to utilise intuitive thinking and decision-making.
The empirical evidence on the factors influencing the participants’ preference for the
various risk management techniques have been addressed in Interview Question 11.
376 Appendix D Case 4 Data Presentation
Interview Question 11 What are the factors influencing your preference of the above risk
management technique?
1st ranked technique ……………………………………………………………………………
2nd ranked technique ……………………………………………………………………………
3rd ranked technique ……………………………………………………………………………
Findings: The findings have been presented in Table D.11.
Case study 4
Risk register Good for risk allocation and as an audit trail for checking risk 1
management responses
Simple way of recording and presenting project risk from 1
project inception
Familiar method of risk identification and treatment 3
Personal experience Opportunity to draw from lessons learnt on previous projects 4
Brainstorming Effective in using your previous experience in risk identification 1
Opportunity to utilise knowledge from all the project team 4
rather than just one person
Gut feeling Use your experience in risk identification and treatment 2
Theoretical Issue 7
– Theoretical issue 7: Psychometric research findings suggest that most people relate better
to narratives, models, illustrations, and anecdotal simulations better than statistical data
(Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals preference for quali-
tative data presentation formats, over statistical data, and quantitative models.
The related interview questions are 12–14 (see Appendix A). The study data has been
examined for evidence on preference for construction risk management data presentation
format. The interview protocol was designed to stimulate intuitive thinking and decision
processing by allowing the participants limited time to respond to the questions.
Findings: The findings from case study 4 have been summarised below in Figures D.10–
D.17.
Interview Question 12 Which of the following forms of risk data presentations do you
know? (Please tick all the applicable)
Narrative …… . . . .. Probabilities ……………… . . . .
Qualitative statements ………… Whole numbers ………… . . . .
Graphs ……… . . . .. Ratios ……………………………
Fractions ……… . . . . Scenarios …… . . . .
Simulations ……… . . . .. Other (please specify) ……………
D.3 Research Proposition 3: Findings 377
Findings: The responses against the risk management data presentation formats have
been presented in Figure D.10. Figure D.11 also categorises the responses into qualitative
and quantitative presentation formats.
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00% 18.00%
66.67%
Qualitative Quantitative
Interview Question 13 Which of the following forms of risk data presentations have you
applied on this project? (Please tick all the applicable).
Narrative …… . . . .. Probabilities ……………… . . . .
Qualitative statements ………… Whole numbers ………… . . . .
Graphs ……… . . . .. Ratios ……………………………
Fractions ……… . . . . Scenarios …… . . . .
Simulations ……… . . . .. Other (please specify) ……………
378 Appendix D Case 4 Data Presentation
Findings: The responses against the risk management data presentation formats have
been presented in Figure D.12. Figure D.13 also categorises the responses into qualitative
and quantitative presentation formats.
Ratios – Quantitative 0%
Simulations – Qualitative 4%
Fractions – Quantitative 0%
Graphs – Quantitative 4%
80.00%
Qualitative Quantitative
Interview Question 14 On a scale of 1–3, please rank the top three (3) risk data presentation
formats which you would prefer to use on your future risk management practices? You may
select from the following list – Narratives, Probabilities, Qualitative statements, Whole num-
bers, Graphs, Ratios, Fractions, Scenarios, and Simulations (please note: A scale of 1 represents
the highest preference, and a scale of 3 represents the lowest preference).
D.3 Research Proposition 3: Findings 379
Findings: The responses against the risk management data presentation formats have
been presented in Figure D.14. Figure D.15 also categorises the responses into qualitative
and quantitative presentation formats.
72.13%
Qualitative Quantitative
380 Appendix D Case 4 Data Presentation
A trend analysis of the responses on the risk management data presentation formats
known, applied, and preferred have been carried out to see if there are any pattern in the
responses that could shed light on the rationale behind the present risk management prac-
tices. The evidence has been presented below in Figures D.16 and D.17.
Figure D.16 Trend analysis of the risk management data presentation formats known, applied,
and preferred.
80.00% 76.00%
72.13%
66.67%
70.00%
60.00%
50.00%
40.00% 33.33%
27.87%
30.00% 24.00%
20.00%
10.00%
0.00%
Known Applied Prefered
Qualitative 66.67% 76.00% 72.13%
Quantitative 33.33% 24.00% 27.87%
Figure D.17 Trend analysis of the risk management data presentation formats known, applied,
and preferred (categorisation into qualitative and quantitative).
D.3 Research Proposition 3: Findings 381
The evidence from the trend analysis has revealed high responses for qualitative data
presentation formats across the three assessment variables. This suggests the study
participants’ current risk management practices of qualitative data processing may
continue for some time.
A major concern however is the variance between the proportions of applied systems
of risk management techniques presented under Theoretical Issue 6 (87.5% intuitive ver-
sus 12.5% rational) and the applied systems of risk data formats (76% qualitative versus
24% quantitative), which suggests theoretically incompatible practices of intuitive process-
ing of some quantitative data formats. This was validated in the responses of the contracts
manager and project manager, who after confirming sole reliance on intuitive techniques
under Theoretical Issue 6, have also admitted quantitative probability data application. The
previous theoretical discussion on the need for complementarity between a decision pro-
cessing system and its data presentation format (Kahneman 2011) suggests the likelihood
of psychological conflicts and reduced accuracy in their ensuing risk analysis.
Theoretical Issue 8
– Theoretical issue 8: Most people do not trust, understand, or use statistics and probability
estimates in their risk management decision-making (March and Shapira 1987; Akintoye
and Macleod 1997; Edwards and Bowen 1998; Lyons and Skitmore 2004).
Theoretical issue to be confirmed if the empirical evidence reveals low level of knowledge
and application of statistics and probability.
The related interview questions are 15–18 (see Appendix A). The study data have been
examined for evidence on the participants’ level of knowledge and understanding of statis-
tics and probability assessment. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions intuitively without consulting any reference source.
Findings: The findings from case study 4 have been summarised below in Table D.12 and
Figures D.18–D.20.
Interview Question 15 Which of the following statistical symbols are you familiar with?
Please indicate by selecting from the following list: summation, probability density
function, probability function, sample mean, population mean, median, variance, standard
deviation, sample standard deviation, don’t know.
Symbol Name
P(A)
S
∑
Case study 4
Statistical No. of correct
symbol responses Percentages (%)
Interview Question 16–18: Please use the hypothetical risk evaluations represented in the
table below to answer questions 16–18.
Probability of Probability of
Risk event occurrence impact
A 1/7 30/35
B 97.9% 84%
C 1 : 9743 4:5
57.14% 42.86%
Interview Question 17 Which of the above risk events is certain not to occur?
Risk event A ……………………………………… . . . ..
Risk event B ……………………………………… . . . ..
Risk event C ……………………………………… . . . ..
All (risk events A, B, and C) …………………… . . . ..
None of the risk events ………………………… . . . ..
Don’t know ……………………………………………
Findings: The percentage of correct responses has been presented in Figure D.19.
57.14% 42.86%
Interview Question 18 Which of the above risk events is certain to have the highest impact?
Risk event A ……………………………………… . . . ..
Risk event B ……………………………………… . . . ..
Risk event C ……………………………………… . . . ..
All (risk events A, B, and C) ………………………
None of the risk events …………………… . . . ..
Don’t know ……………………………………… . . . ..
Findings: The percentage of correct responses has been presented in Figure D.20.
100.00%
The empirical evidence has revealed significant low level of understanding of statistical
symbols and probability assessments among the case study participants:
– 3 out of 28 correct identification of statistical symbols which represents 10.71% (see
Table D.12).
– 6 out of 21 correct responses for the probability assessments which represents 28.57%
(see page 5 of Appendix D). A significant empirical discovery is the low accuracy in
responses from the project manager and the contracts manager, which appears consis-
tent with the previous predication of reduced precision in their risk analysis output (see
Theoretical Issue 7), due to their theoretically incompatible risk management practice
of intuitive processing of quantitative probability data. The project manager’s responses
were all wrong, whilst the contracts manager also gave one wrong response. The overall
responses from the case reflected erroneous linkage of higher probability prediction to
certainty of occurrence of an event, which may also have resulted from a general lack of
understanding of the probability concept.
Theoretical Issue 9
– Theoretical issue 9: Most people relate better to qualitative assessments than quantitative
assessments (Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals high accuracy for risk
analysis expressed in whole numbers, qualitative statements, and scenarios, compared to
risk analysis expressed in percentages, decimals, and ratios.
The related interview questions are Interview Question 19 and Interview Question 20
(see Appendix A). The study data has been examined for evidence on the participants’
level of knowledge and understanding of qualitative and quantitative assessments. The
interview protocol was designed to stimulate intuitive thinking and decision processing by
allowing the participants limited time to respond to the questions intuitively without using
calculators.
Findings: The findings from case study 4 have been summarised below in Figures D.21
and D.22, and Table D.13.
Interview Question 19 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. There is a 75.2% probability of half of the site being contaminated (… . . . .)
B. There is a 1 : 321 probability of the site being contaminated (… . . . .)
C. There is a 0.381 probability of the site being contaminated (… . . . .)
Findings: The percentage of correct responses has been presented in Figure D.21.
Interview Question 20 Which of the following hypothetical project settings offer the high-
est probability estimate? (Indicate with a tick)
A. The probability of projects over running their initial budget is 3 out of 10 (… . . . .)
B. out of every projects overrun their initial budget (… . . . .)
C. After reviewing the financial records of the four completed construction projects, only
project A recorded a final cost higher than the initial budget estimate (… . . . .)
Findings: The percentage of correct responses has been presented in Figure D.22.
D.3 Research Proposition 3: Findings 385
71.43%
71.43%
The empirical evidence has revealed higher percentage of accuracy for the qualitative
probability equations compared to the quantitative probability equations:
– Qualitative probability equation: Five out of seven correct responses representing 71.43%.
– Quantitative probability equation: Two out of seven correct responses representing
28.57%. Another significant evidence is the composition of the wrong responses. All
the five wrong respondents selected the probability equation expressed in percentages.
Further questioning on the rationale behind the preference for the ‘percentage equation’
format revealed the following responses:
a. Percentages are what we are familiar with from our primary and secondary educa-
tion – two responses.
b. Percentages are what we use in our everyday working life – two responses.
c. Percentages are easy to understand – one response.
The above responses suggest that the participants who selected the percentage equation
were simply relying on their availability heuristics to intuitively generate the answers.
There were two further supplementary questions to solicit empirical evidence on the par-
ticipants’ experience with answering the two forms of probability equations. The findings
have been presented below in Table D.13. The empirical evidence has revealed the qualita-
tive equation format to be comparatively straightforward and easy to understand.
386 Appendix D Case 4 Data Presentation
Table D.13 Participants’ experience with answering the quantitative and qualitative probability
equations.
Assessment Percentages
Supplementary question format Response (%)
Theoretical Issue 10
– Theoretical issue 10: Where a potential outcome of a future event evokes a strong
affective memory, the assessor may be insensitive to the probability predictions on the
event (Slovic et al. 2010).
Theoretical issue to be confirmed if the empirical evidence reveals insensitivity to differ-
ent ranges of hypothetical probability predictions of asbestos contamination risk.
The related interview questions are 21i–21iii (see Appendix A). The study data has been
examined for evidence on the participants’ responses to the different range of hypothetical
asbestos contamination risk. The interview protocol was designed to stimulate intuitive
thinking and decision processing by allowing the participants limited time to respond to
the questions.
Findings: The findings from case study 4 have been summarised below in Table D.14.
Interview Questions 21
Asbestos related illness has been a major cause of construction occupational death. Accord-
ing to the Health and Safety Executive in the United Kingdom, there has been a recent surge
in asbestos related illness and deaths (see the table and pictures below).
Asbestos related deaths
1968 153
2009 2321 411
2010 2347 412
2011 2291 429
2012 2535 464
Source: Shutterstock.com
Picture 2: Asbestos removal by licenced contractors
21i Please kindly review the information above and complete the ‘Appropriate risk
response’ column of the ‘Hypothetical project risk evaluation matrix’ below. Please select
your response from any of the following options:
i. Appoint a licenced contractor to carry out further surveys to identify and remove the
asbestos before the commencement of construction works.
ii. Carry on construction works at the non-contaminated area whilst carrying out further
investigations at the remaining areas to identify and remove any asbestos residue.
iii. Supply protective masks to all operatives and carry on with the construction works.
iv. Carry on with the construction works and forget about any asbestos.
Hypothetical project risk evaluations matrix
21ii What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
(Please select your response from any of the options given under 21i.)
21iii What will be your risk response if the risk evaluation results were to change to:
Hypothetical project risk evaluations matrix
(Please select your response from any of the options given under 21i.)
Findings: The participants’ risk responses under the different probability predictions have
been presented in Table D.14.
Table D.14 The impact of probability predictions on events which evokes strong affective memories.
21i 95% probability of 99% probability of death Appoint a licenced contractor to carry 7 out of 7 (100%)
mesothelioma causing resulting from exposure out further surveys to identify and remove
asbestos fibre contamination the asbestos before the commencement of
on 100% of the site construction works
21ii 10% probability of 50% probability of asbestos Appoint a licenced contractor to carry 6 out of 7 (85.71%)
non-malignant asbestos infection out further surveys to identify and remove
contamination on 10% of the the asbestos before the commencement of
site construction works
Carry on construction works at the 1 out of 7 (14.29%)
non-contaminated area whilst carrying
out further investigations at the
remaining areas to identify and remove
any asbestos residue
21iii 1% probability of 1% probability of Appoint a licenced contractor to carry 4 out of 7 (57.14%)
non-malignant asbestos non-malignant pleural out further surveys to identify and remove
contamination within a infection the asbestos before the commencement of
specific 1% area of the site construction works
Carry on construction works at the 3 out of 7 (42.86%)
non-contaminated area whilst carrying
out further investigations at the
remaining areas to identify and remove
any asbestos residue
390 Appendix D Case 4 Data Presentation
The empirical evidence has revealed significant insensitivity to the different ranges of
hypothetical probability predictions on asbestos contamination. About 57.14% of the study
participants consistently responded that they would appoint a licensed contractor to carry
out further asbestos surveys to identify and remove asbestos contaminated materials before
they commence site construction works. The different levels of probability risk predictions
could not alter the responses from the technical manager, project architect, client devel-
opment manager, and project engineer, not even for the very low probability prediction of
non-malignant contamination within a small specified area.
The empirical evidence from the direct observations also revealed some level of emotional
discomfort for some of the participants during the interrogations, especially the statistical
and probability assessments (Theoretical Issue 8) and the quantitative equations (Theo-
retical Issue 9). Some of the participants even inquired if they could search the statistical
symbols on the internet and process the quantitative equations with calculators. When this
was denied, the frustration became evident in their body language, confirming the psy-
chological conflict they were facing in attempting intuitive processing of quantitative data
formats.
In summary, the empirical evidence from Theoretical Issues 6–10 has provided sub-
stantial support for research proposition 3. The empirical findings have revealed low level
training in formal risk management techniques, and intuitive dominated risk management
practices among the study participants. There has also been evidence of immense qualita-
tive data application and preference. The empirical findings have also revealed disparities
between the proportions of applied systems of risk management approach versus the
applied data processing formats, with evidence of intuitive processing of quantitative
probability by the project manager and contracts manager. All the differentiated specialist
roles exhibited low proficiency in statistical symbols, and probability judgement. The
project manager in particular performed poorly in the probability judgement, thereby
validating the analytical effects of his theoretically incompatible data processing practices.
A review of the study participants’ risk assessment has further revealed comparatively
higher accuracy for intuitive evaluation of qualitative risk data formats over quantitative
risk data formats; and insensitivity to probability predictions of emotive events. Bringing all
the above together, confirms the incompatibility of mixing decision processing approaches
and data presentation formats from the different decision-making systems.
391
References
Abderisak, A. and Lindahl, G. (2015). Take a chance on me? Construction client’s perspectives
on risk management. Procedia Economics and Finance 21: 548–554.
Adams, F.K. (2008). Risk perception and Bayesian analysis of international construction
contract risks: the case of payment delays in a developing economy. International Journal of
Project Management 26: 138–148.
Akintoye, A.S. and Macleod, M.J. (1997). Risk analysis and management in construction.
International Journal of Project Management 15 (1): 31–38.
Alhakami, A.S. and Slovic, P. (1994). A psychological study of the inverse relationship between
perceived risk and perceived benefit. Risk Analysis 14 (6): 1085–1096.
Allcoat, J. (2018). Construction Statistics, Great Britain, 2018. Office of National Statistics.
https://www.ons.gov.uk/businessindustryandtrade/constructionindustry/articles/
constructionstatistics/2018 (accessed on 9 February 2020).
Allinson, K. (2008). Architects and Architectures of London. Oxford: Elsevier.
Arthur, A.C. (2017). Is It Time to Learn the Lessons? Fire Incidence at Atomic Junction.
Ghanaian Daily Graphic, October 16, 2017. No. 20510, p. 7.
Arthur, A.C. (2018). Do Construction Project Managers Need Training in Behavioural Sciences
to Effectively Manage Project Risk? RICS COBRA 2018 (23–24 April 2018). London, UK:
RICS HQ.
Arthur, A.C. (2020). Linkages, networks, and interactions: exploring the context for risk
decision making in construction supply chains. In: Successful Construction Supply Chain
Management; Concepts and Case Studies (ed. S. Pryke), 143–165. Wiley-Blackwell. ISBN:
978-1-119-45054-2.
Arthur, A.C. and Pryke, S.D. (2013). Instinctive thinking; analysis of construction risk
management systems. Proceedings of the Construction, Building and Real Estate Research
Conference of the Royal Institution of Chartered Surveyors, New Delhi, September 2013.
London: RICS.
Arthur, A.C. and Pryke, S.D. (2014). Are we adding risk to our projects by mixing objective
assessments of compound conjunctive and disjunctive project risks with intuitive
approaches? In: Proceedings 30th Annual ARCOM Conference, 1–3 September 2014
(eds. A.B. Raiden and E. Aboagye-Nimo), 1399–1408. Portsmouth: Association of
Researchers in Construction Management.
Aven, T. (2010). Misconceptions of Risk. Chichester: Wiley.
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
392 References
Carnevale, P.J.D. and Isen, A.M. (1986). The influence of positive affect and visual access on the
discovery of integrative solutions in bilateral negotiation. Organizational Behavior and
Human Decision Processes 37: 1–13.
CBI (2010). Procuring in a Downturn. CBI.
Chabris, C. and Simons, D. (2010). The Invisible Gorilla: And Other Ways Our Intuitions Deceive
Us. Random House, NY: Crown Publishers.
Chapman, R.J. (1999). The likelihood and impact of changes of key project personnel on the
design process. Construction Management and Economies 17: 99–106.
Chapman, G.B. and Johnson, E.J. (2002). Incorporating the irrelevant: anchors in judgments of
belief and value. In: Kahneman Heuristics and Biases, The Psychology of Intuitive Judgment
(eds. T. Gilovich and D. Griffin), 120–138. Cambridge: Cambridge University Press.
Chauvin, B., Hermand, D., and Mullet, E. (2007). Risk perception and personality facets. Risk
Analysis 27 (1): 171–185.
Checkland, P. (1999). Systems Thinking, Systems Practice. Chichester: Wiley.
Cohen, J., Chesnick, E.I., and Haran, D.A. (1972). A confirmation of the INERTIAL-ψ effect in
sequential choice and decision. British Journal of Psychology 63 (1): 41–46.
Cole, G.A. (2004). Management Theory and Practice, 27. Singapore: Seng Lee Press.
Cole, G. and Kelly, P. (2020). Management and Practice, 9e. Cengage Learning EMEA.
Cova, B. and Salle, R. (2006). Communications and stakeholders. In: The Management of
Complex Projects, A Relationship Approach (eds. S. Pryke and H. Smyth), 131–146. Oxford:
Blackwell Publishing.
Cox, E. (1999). The Fuzzy Systems Handbook: A Practitioner’s Guide to Building, Using, and
Maintaining Fuzzy Systems, 2e. London: AP Professional.
Creswell, J.W. (2013). Qualitative Inquiry & Research Design: Choosing Among Five Approaches,
3e. London: Sage.
Dallas, M. (2006). Value and Risk Management, A Guide to Best Practice. Oxford: Blackwell
Publishing Ltd.
Damasio, A. (2006). Descartes’ Error, Emotion, Reason and the Human Brain. London: Vintage
Books.
Davies, A. (2018). Modern Methods of Construction, A Forward-Thinking Solution to the Housing
Crisis? Royal Institution of Chartered Surveyors.
Denes-Raj, V. and Epstein, S. (1994). Conflict between intuitive and rational processing: when
people behave against their better judgment. Journal of Personality and Social Psychology 66
(5): 819–829.
Denzin, N.K. and Lincoln, Y.S. (2007). Introduction: the discipline and practice of qualitative
research. http://www.sagepub.com/sites/default/files/upm-binaries/17670_Chapter1.pdf
(accessed on 12 August 2015).
Denzin, N.K. and Lincoln, Y.S. (2011). Introduction: The Discipline and Practice of Qualitative
Research. The Sage Handbook of Qualitative Research, 4e. Thousand Oaks, CA: Sage.
Derry, T.K. and Williams, T.I. (1960). A Short History of Technology, From The Earliest Times to
A.D. 1900. Oxford University Press.
Dohle, S., Keller, C., and Siegrist, M. (2010). Examining the relationship between affect and
implicit associations: implications for risk perception. Risk Analysis 30 (7): 1116–1128.
394 References
Edkins, A. (2009). Risk management and the supply chain. In: Construction Supply Chain
Management: Concepts and Case Studies (ed. S. Pryke), 115–136. Oxford: Blackwell
Publishing.
Edwards, P.J. and Bowen, P.A. (1998). Risk and risk management in construction: a review and
future directions for research. Engineering Construction and Architectural Management 5 (4):
339–349.
Egan, S.J. (1998). Rethinking Construction - The Report of the Construction Task Force. London:
Department of Trade and Industry.
Eisenhardt, K.M. (1989). Building theory from case study research. The Academy of
Management Review 14 (4): 532–550.
Epley, N. and Gilovich, T. (2002). Putting adjustment back in the and adjustment heuristic. In:
Kahneman Heuristics and Biases, The Psychology of Intuitive Judgment (eds. T. Gilovich and
D. Griffin), 139–149. Cambridge: Cambridge University Press.
Everis (2009). Risk Management in the Insurance Business Sector. MFC Artes Graficas, S.L.
http://www.everis.com/global/WCRepositoryFiles/Study%20of%20Risk%20Management
%20-%20sept%2009.pdf (accessed on 29 December 2015).
Fellows, R. and Liu, A. (2008). Research Methods for Construction, 3e. Chichester: Blackwell
Publishing Ltd.
Finucane, M.L., Peters, E., and Slovic, P. (2003). Judgement and decision making: the dance of
affect and reason. In: Emerging Perspectives on Judgement and Decision Research (eds.
S.L. Schneider and J. Shanteau), 327–346. Cambridge: Cambridge University Press.
Fischhoff, B., Slovic, P., Lichtenstein, S. et al. (1978). How safe is safe enough? A psychometric
study of attitudes towards technological risks and benefits. Policy Sciences 9 (2): 127–152.
Flyvbjerg, B., Bruzelius, N., and Rothengatter, W. (2003). Megaprojects and Risk, An Anatomy of
Ambition. Cambridge: Cambridge University Press.
Forgas, J.P. (1995). Mood and judgement: the affect infusion model (AIM). Psychology Bulletin
117 (1): 39–66.
Forgas, J.P. (2001). Feeling and Thinking: The Role of Affect in Social Cognition. Cambridge:
Cambridge University Press.
Forgas, J.P. (2002). Feeling and doing: affective influences on interpersonal behavior.
Psychological Inquiry 13 (1): 1–28.
Garratt, M. (2010). A four pronged approach. Construction Journal, RICS April–May 2010.
Gate Safe (2020). Making Gates Safe. https://gate-safe.org/ (accessed on 23 December 2020).
Gov.UK (2021a). Check a public building’s Display Energy Certificate. https://www.gov.uk/
check-energy-performance-public-building (accessed on 8 February 2021).
Gov.UK (2021b). Energy Performance Certificates for your business premises. https://www.gov
.uk/energy-performance-certificate-commercial-property (accessed on 8 February 2021).
Gov.UK (2021c). Find an energy assessor. https://www.gov.uk/find-an-energy-assessor
(accessed on 8 February 2021).
Gov.UK (2021d). Building a new home and VAT. https://www.gov.uk/vat-building-new-home
(accessed on 8 February 2021).
Gov.UK (2021e). Stamp duty land tax, reliefs and exemptions. https://www.gov.uk/stamp-
duty-land-tax/reliefs-and-exemptions (accessed on 13 February 2021).
References 395
Ive, G.J. and Gruneberg, S.L. (2000). The Economies of the Modern Construction Sector.
Basingstoke: Macmillan Press Ltd.
Jain, L.C. and Martin, N.M. (1998). Fusion of Neural Networks, Fuzzy Sets and Genetic
Algorithms- Industrial Application. London: CRC Press.
Japp, K. and Kusche, I. (2008). Systems theory and risk. In: Social Theories of Risk and
Uncertainty: An Introduction (ed. J.O. Zinn), 52–75. Oxford: Blackwell Publishing.
JCT News (2012). Low-Carbon Building That Does Not Cost the Earth. London: JCT. Sweet &
Maxwell.
Kahneman, D. (2011). Thinking Fast and Slow. London: Penguin Books Ltd.
Kahneman, D. and Tversky, A. (1973). On the psychology of prediction. Psychological Review 80
(4): 237–251.
Kahneman, D. and Tversky, A. (1979). Prospect theory: an analysis of decision under risk.
Econometrica 47 (2): 263–292.
Kahneman, D. and Tversky, A. (1982a). Subjective probability: a judgment of
representativeness. In: Tversky Judgement Under Uncertainty: Heuristics and Biases (eds. D.
Kahneman, P. Slovic and A. Tversky), 32–47. Cambridge: Cambridge University Press.
Kahneman, D. and Tversky, A. (1982b). On the psychology of prediction. In: Judgement Under
Uncertainty: Heuristics and Biases (eds. D. Kahneman, P. Slovic and A. Tversky), 48–68.
Cambridge: Cambridge University Press.
Kamper, E. (2000). Decision Making Under Risk in Organisationals; The Case of German Waste
Management. Aldershot: Ashgate Publishing Limited.
Khazaeni, G., Khanzadi, M., and Afshar, A. (2012a). Fuzzy adaptive decision making model for
selection balanced risk allocation. International Journal of Project Management 30: 511–522.
Khazaeni, G., Khanzadi, M., and Afshar, A. (2012b). Optimum risk allocation model for
construction contracts: fuzzy TOPSIS approach. Canadian Journal of Civil Engineering 39:
789–800.
Knight, A. and Turnbull, N. (2008). Epistemology. In: Advanced Research Methods in the Built
Environment (eds. A. Knight and L. Ruddock). Chichester: Blackwell Publishing Ltd.
Kululanga, G. and Kuotcha, W. (2010). Measuring project risk management process for
construction contractors with statement indicators linked to numerical scores. Engineering
Construction and Architectural Management 17 (4): 336–351.
Kuo, Y. and Lu, S. (2013). Using fuzzy multiple criteria decision making approach to enhance
risk assessment for metropolitan construction projects. Internal Journal of Project
Management 31: 602–614.
Larsen, R.J. and Diener, E. (1987). Affect intensity as an individual difference characteristic: a
review. Journal of Research in Personality 21: 1–39.
Latham, S.M. (1994). Constructing the Team: Joint Review of Procurement and Contractual
Arrangements in the United Kingdom Construction Industry. Final Report. London: HMSO.
ISBN 0-11-752994-X, https://constructingexcellence.org.uk/wp-content/uploads/2014/10/
Constructing-the-team-The-Latham-Report.pdf.
Lock, D. (2003). Project Management, 8e. Hampshire: Gower Publishing Limited.
Loosemore, M. (2006). Managing project risks. In: The Management of Complex Projects, A
Relationship Approach (eds. S. Pryke and H. Smyth), 131–146. Oxford: Blackwell Publishing.
Loosemore, M., Raftery, J., Reilly, C., and Higgon, D. (2006). Risk Management in Projects.
Oxon: Taylor & Francis.
References 397
Slovic, P. (2010). The Feeling of Risk, New Perspective on Risk Perception. London: Earthscan Ltd.
Slovic, P. and Lichtenstein, S. (1983). Preference reversal: a broader perspective. American
Economic Review 73 (4): 596–605.
Slovic, P. and Peters, E. (2006). Risk perception and affect. Currrent Directions in Psychological
Science 15 (6): 322–325.
Slovic, P., Fischhoff, B., Lichtenstein, S., and Roe, F.J.C. (1981). Perceived risk: psychological
factors and social implications and discussion. Proceedings of the Royal Society of London.
Series A, Mathematical and Physical Sciences 376 (1764): The Assessment and Perception of
Risk, 17–34.
Slovic, P., Fischhoff, B., and Lichtenstein, S. (1982). Facts versus fears: Understanding
perceived risk. In: Judgement Under Uncertainty: Heuristics and Biases (eds. D. Kahneman,
P. Slovic and A. Tversky), 463–489. Cambridge: Cambridge University Press.
Slovic, P., Finucane, M., Peters, E., and MacGregor, D.G. (2002). The affect heuristic. In:
Heuristics and Biases, The Psychology of Intuitive Judgment (eds. T. Gilovich, D. Griffin and
D. Kahneman), 397–420. Cambridge: Cambridge University Press.
Slovic, P., Finucane, M.L., Peters, E., and MacGregor, D.G. (2010). Risk as analysis and risk as
feelings: some thoughts about affect, reason, risk and rationality. In: The Feeling of Risk, New
Perspective on Risk perception (ed. P. Slovic), 21–36. London: Earthscan Ltd.
Smith, N.J., Merna, T., and Jobling, P. (2006). Managing Risk in Construction Projects, 2e.
Oxford: Blackwell Publishing.
Stanovich, K.E., West, R.F., and Toplak, M.E. (2016). The Rationality Quotient, Toward a Test of
Rational Thinking. Cambridge, MA: First MIT Press.
Staw, B.M. and Barsade, S.G. (1993). Affect and managerial performance: a test of the
sadder-but-wiser vs. happier-and-smarter hypotheses. Administrative Science Quarterly 38
(2): 304–331.
Sutrisna, M. (2009). Research methodology in doctoral research: understanding the meaning of
conducting qualitative research, working paper. In: Proceedings of the Association of
Researchers in Construction Management (ARCOM) Doctoral Workshop (ed. A. Ross), 48–57.
Liverpool. ISBN: 978-0-9562622-0-8.
The Economic Times (2011). “Blackberry riots” in Britain. Addition to Smart phones. http://
economictimes.indiatimes.com/news/international-business/blackberry-riots-in-britain-
addiction-to-smartphones/articleshow/9601104.cms (accessed on 14 February 2021).
Thompson, M., Ellis, R., and Wildavsky, A. (1990). Culture Theory. Oxford: Westview Press.
Tulloch, J. (2008). Culture and risk. In: Social Theories of Risk and Uncertainty: An Introduction
(ed. J.O. Zinn), 52–75. Oxford: Blackwell Publishing.
Tversky, A. and Kahneman, D. (1982a). Judgement under uncertainty: heuristics and biases.
In: Judgement Under Uncertainty: Heuristics and Biases (eds. D. Kahneman, P. Slovic and
A. Tversky), 3–20. Cambridge: Cambridge University Press.
Tversky, A. and Kahneman, D. (1982b). Belief in the law of small numbers. In: Judgement
Under Uncertainty: Heuristics and Biases (eds. D. Kahneman, P. Slovic and A. Tversky),
23–31. Cambridge: Cambridge University Press.
Tversky, A. and Kahneman, D. (1982c). Availability: a heuristic for judging frequency and
probability. In: Judgement Under Uncertainty: Heuristics and Biases (eds. D. Kahneman,
P. Slovic and A. Tversky), 163–178. Cambridge: Cambridge University Press.
400 References
Tversky, A. and Kahneman, D. (2002). Extensional versus intuitive reasoning: the conjunction
fallacy in probability judgment. In: Heuristics and Biases, The Psychology of Intuitive
Judgment (eds. T. Gilovich, D. Griffin and D. Kahneman), 19–48. Cambridge: Cambridge
University Press.
UNEP (2021). About UN Environment Programme. https://www.unep.org/about-un-
environment (accessed on 8 February 2021).
Walker, A. (2007). Project Management in Construction, 5e. Oxford: Blackwell Publishing Ltd.
Walker, A. (2015). Project Management in Construction, 6e. Chichester: Wiley Blackwell.
Weber, E.U. and Milliman, R.A. (1997). Perceived risk attitudes; relating risk perception to
risky choice. Management Science 43 (2): 123–144.
Wetherbe, J.C. and Vitalari, N.P. (1994). Systems Analysis and Design; Best Practice, 4e. New
York: West Publishing Company.
Winch, G.M. (2010). Managing Construction Projects, 2e. Chichester: Blackwell Publishing Ltd.
Yin, R.K. (2003). Case Study Research; Design and Methods, 3e. London: SAGE Publications Ltd.
Yin, R.K. (2014). Case Study Research; Design and Methods, 5e. London: SAGE Publications Ltd.
Zajonc, R.B. (1980). Feeling and thinking preferences need no inferences. American
Psychologist 35 (2): 151–175.
Zajonc, R.B. (2001). Mere exposure: a gateway to the subliminal. Current Directions in
Psychological Science 10 (6): 224–228.
Zeng, J., An, M., and Smith, N.J. (2007). Application of a fuzzy based decision making
methodology to construction project risk assessment. International Journal of Project
Management 25: 589–600.
Zinn, J.O. (2008). Social Theories of Risk and Uncertainty An Introduction. Oxford: Blackwell
Publishing.
401
Index
Construction Risk Management Decision Making: Understanding Current Practices, First Edition. Alex C. Arthur.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
402 Index
230–233, 236, 240, 245, 249, 252, 255, construction period 9, 10, 29
280, 281, 283, 285, 287, 288, 290, 293, construction products 1, 9–11, 13, 27–29
296, 300, 305, 309, 312, 338, 339, 341, construction project delivery 1–3, 56, 74, 83,
342, 344, 346, 348, 351, 357, 362, 366, 84, 91, 102, 146, 148, 153, 157, 161,
369 168, 169, 196, 202, 205, 206, 225, 232,
component 3, 4, 9, 11, 13, 14, 17, 21, 29, 65, 279, 293, 335, 348
66, 83, 84, 91, 97, 162, 163, 168, 196, construction system 9, 10, 12–14, 22, 25–30
199, 202, 207, 208, 225, 229, 279, 282, constructor 23, 25, 115, 221, 245, 301
335, 338 contemporary 1–3, 11, 23, 25, 30, 31, 35, 36,
concept 2, 3, 5, 7, 9–11, 14, 16–18, 29–33, 38, 41, 59, 60, 65, 68, 71, 72, 182
35–39, 41, 43, 49, 65, 69, 71, 72, 86, 87, contract 18, 25, 28, 41, 42, 45, 49, 72, 89, 90,
89, 93, 95, 96, 109, 139, 154, 161, 168, 93, 94, 148, 151, 155, 179, 198, 203, 352
186, 195, 196, 199, 201–204, 227, 239, contractor 13–16, 25, 29, 42, 47, 50, 90, 92,
270, 281, 295, 327, 334, 337, 384 93, 96, 101, 109, 142, 144, 145, 147–149,
conceptual 2, 7, 9, 17, 20, 30, 31, 33–39, 49, 151, 152, 155, 156, 183, 186, 192, 194,
52, 60, 65, 68–72, 77, 83, 86, 88, 169, 212, 218, 221, 228, 233, 236, 240, 252,
185, 196, 201, 205 261, 273, 275, 276, 287, 331–333, 339,
construction 1–4, 7, 9–30, 32, 35, 37–39, 41, 342, 351, 356, 362, 387–390
43–52, 56–62, 64–74, 77, 78, 81–87, contracts manager 2, 89, 90, 94, 98, 100, 124,
89–98, 100, 102, 106, 107, 111, 112, 147, 148, 150, 151, 154, 155, 157, 158,
114, 116, 117, 120, 123–125, 129, 130, 162, 163, 171, 174, 175, 179, 181, 183,
132, 134, 140, 142, 144–153, 155–159, 185–190, 193, 195, 198, 199, 212, 226,
161–165, 167–170, 173, 175, 176, 247, 255, 261, 270, 276, 280, 288, 317,
178–180, 182, 183, 185, 186, 189, 192, 324, 327, 334, 336, 341, 369, 384, 390
194–216, 218–223, 225–230, 232, 234, control 2, 11, 17–19, 25, 26, 42, 43, 45, 46,
237, 238, 240, 242, 243, 245–247, 250, 59, 68, 93, 96, 109, 339, 351
253, 255, 256, 261, 270–273, 275, 276, Corinthians 11
279–286, 288, 292–294, 296–298, 300, coronavirus 25
302, 303, 307, 311–313, 319, 328, 330, corporate activities 1
332, 333, 335–341, 343, 344, 347–349, Covid-19 pandemic 1, 31, 32, 35–38, 42
351, 354, 357, 359, 360, 364, 367, 369, cross-purpose working 10, 19, 21, 24, 30,
370, 376, 384, 386, 388–390 197, 205
construction activities 13, 16, 25 cultural studies 34, 35
construction client 2 culture theory 3, 4, 66, 70, 87, 162, 168, 196,
construction delivery 1, 9–11, 14, 15, 17, 18, 202
20–22, 26, 29, 32, 65–67, 81, 84, 102,
146, 153, 176, 198, 232, 293 d
construction industry 7, 9, 10, 12, 14, 15, 17, data 3, 4, 7, 44, 46, 48–51, 53, 54, 60–63, 68,
19, 21–23, 26, 28–30, 37, 39, 43, 44, 69, 72, 73, 77–79, 81, 82, 84–90, 107,
47–51, 61, 62, 69, 71, 81–83, 87, 90, 92, 112, 117, 124, 125, 132–137, 139, 141,
94, 130, 146, 149, 152, 153, 156, 145–147, 150, 151, 153–155, 157–159,
162–164, 170, 180, 182, 185, 211, 280, 179, 180, 184–186, 189, 192, 195,
336, 341 198–201, 204–211, 214–216, 225, 230,
construction management 3, 10, 11, 41, 50, 234, 238, 247, 255, 256, 261–267, 270,
70, 73, 201, 202 272, 273, 277, 279, 286, 294, 298, 303,
404 Index
grounded heuristics 91, 107, 112, 117, 124, 328, 330, 332, 333, 338, 340–343, 345,
146, 148, 153, 157, 158, 162, 164, 170, 348, 349, 354, 360, 361, 365, 369, 382,
171, 173, 176, 197, 198, 203, 208, 209, 384, 386, 388–390
226, 234, 238, 243, 247, 280, 294, 298,
303, 312, 327, 336, 341, 349, 360, i
369 identification 2, 3, 31, 41, 43–45, 49, 57, 69,
grouping 22, 24, 66 72, 74, 79, 84, 94, 107, 112, 117, 118,
groups 4, 11, 13, 17, 19, 21, 26–28, 30, 64, 65, 121, 124, 131, 139, 146, 150, 153, 155,
70, 94, 163–165, 175, 183, 185, 208 157, 158, 162, 164, 168–171, 173, 176,
guided morality 1, 27, 31, 38, 71 182, 186, 196–198, 201, 203, 205, 208,
209, 212, 234, 238, 243, 247, 248, 251,
h 262, 270, 280, 294, 298, 303, 304, 308,
hazard 1, 32 312, 319, 327, 336, 341, 348, 349, 354,
heuristics 2, 7, 35, 39, 47, 50–52, 55, 56, 360, 361, 365, 369, 376, 384
59–63, 65, 66, 68, 69, 72, 73, 77, 79, ideologies 2, 11
82–85, 87, 91, 94, 97, 98, 102, 107, 108, illustration 13, 17, 19, 24, 25, 35, 55, 60, 61,
112–114, 116–118, 121, 124–130, 141, 85, 117, 125, 132, 150, 157, 180, 210,
146, 148, 153, 157, 158, 162–165,
243, 255, 261, 267, 313, 319, 360, 370,
168–171, 173, 176, 179, 180, 186, 189,
376
192, 196–199, 201–206, 208–210, 213,
impact 1–4, 7, 17, 21–26, 30, 31, 34–39, 41,
225, 226, 229, 230, 232, 234, 238, 239,
42, 44, 45, 49, 51, 52, 57, 59, 60, 64–66,
243, 244, 247, 248, 251, 255–260, 272,
68–70, 73, 74, 83–85, 87, 91, 93, 97,
279, 280, 282, 286, 293–295, 298, 303,
102, 117, 137, 139, 142, 144, 145, 150,
304, 308, 312–317, 327, 329, 335, 341,
153, 157, 162, 169, 186, 189, 196, 199,
348–350, 354, 356, 358, 360, 361, 365,
205, 208, 209, 211, 215, 216, 218, 219,
369–375, 385
221, 225, 229, 234, 247, 268–270, 273,
high rise buildings 1, 49
275, 276, 279, 281–283, 293, 296, 303,
homeostatic 2
319, 325, 327, 332, 333, 335, 337, 338,
homeostatic perspective 2
348, 351, 360, 382, 383, 388, 389
housing 10, 13, 32, 87, 89, 90, 147, 148, 151,
155, 158 importation 9, 10, 29
human 1, 2, 9–13, 17, 29, 31, 36, 38, 46, 55, industry 2, 7, 9, 10, 12–15, 17, 19, 21–23,
57, 68, 71, 73, 82 26–30, 33, 35, 37–39, 41, 43, 44, 47–51,
human condition 1 56, 61, 62, 69, 71–73, 81–83, 87, 90, 92,
human resource 9 94, 130, 146, 148, 149, 151–153, 155,
human subjectivity 2 156, 162–164, 170, 180, 182, 183, 185,
human survival 1, 38 204, 211, 261, 280, 327, 336, 341
hypothetical 17, 57, 79, 97–99, 101–103, 107, infection 25, 32, 38, 142, 145, 219, 273, 276,
112, 117, 118, 121, 124, 137, 140–142, 332, 333, 388, 389
144–146, 148, 150, 153, 155, 157, 158, information 12, 16, 18, 19, 21, 22, 25, 37, 45,
164, 165, 168–170, 173, 175, 176, 192, 46, 51, 56, 57, 59, 60, 62, 63, 69, 82, 85,
195, 197, 208, 211, 215, 216, 218–220, 87, 91–93, 102, 110, 112–114, 116, 124,
222, 229–235, 238, 243, 247, 248, 251, 144, 146, 148, 149, 152, 153, 156–158,
268, 271–273, 275, 276, 282, 284–287, 162, 169, 173, 197, 209, 218, 221, 226,
289, 294, 298, 303, 304, 308, 312, 325, 234, 238, 243, 244, 247, 275, 280, 281,
Index 407
293, 294, 296, 298–300, 312, 332, 336, 198, 202–205, 218, 221, 223, 233, 235,
339, 348, 351, 354, 355, 369, 388 239, 244, 245, 251, 273, 275, 300, 332,
infrastructure 10, 95, 101, 104, 109, 114, 122, 333, 345, 350, 355, 356, 361, 388, 389
344, 368
innovation 11, 24, 27 j
inputs 9, 12, 14, 17–19, 22, 23, 29, 30, 41, judgement 2, 34, 35, 38, 46, 47, 54, 56–58,
43–45, 49, 65, 71 61–64, 82, 146, 150, 154, 157, 159, 189,
institutions 4, 21, 22, 26–29, 31, 34–36, 38, 204, 211, 277, 334, 390
39, 42, 56, 66, 68, 70, 72–74, 162
insurance 21, 31, 33, 35, 38, 41, 42, 45, 49, k
60, 71, 72, 115, 185, 221, 301 knowledge 22, 33, 41, 57, 61, 69, 72, 73, 81,
interactions 7, 9–12, 14, 17, 20, 25, 26, 28, 86, 91, 125, 131, 137, 139, 162, 180,
30, 31, 34–36, 38, 41, 43, 45, 49, 51, 55, 184, 186–188, 198, 203, 204, 210,
66, 71–73, 78, 86, 163, 221 213–215, 226, 256, 261, 262, 267, 268,
interest 2, 23, 26, 28, 36, 37, 39, 72, 74, 86, 270, 280, 313, 318, 319, 324, 325, 328,
202 336, 370, 375, 376, 381, 382, 384
internal 3, 4, 7, 9, 11, 12, 17, 19, 20, 22–25,
29, 30, 37, 39, 41, 43, 44, 49, 66, 70–72, l
74, 77, 84, 88, 90, 91, 148, 154, legal 10, 26, 28, 30, 42, 44, 71, 93, 94, 98, 100,
161–164, 170, 173, 180, 183, 186, 189,
101, 105, 111, 115, 123, 124, 163, 165,
195, 202, 207, 225, 261, 279, 335
168, 175, 179, 221, 226–228, 230–233,
intuition 41, 46, 47, 50, 51, 55, 61, 63, 64, 66,
237, 241, 246, 249, 253, 255, 280–283,
72, 82, 85, 124, 125, 180, 182, 210, 255,
286, 288, 291, 297, 301, 306, 310, 312,
256, 312, 313, 369, 370
336, 338, 339, 341, 343, 344, 347, 348,
intuitive 3, 4, 7, 41, 46–48, 50–52, 55, 59–64,
352, 358, 363, 367, 369
66, 68, 69, 72–74, 77–79, 81–87, 91,
legal functional subsystem 10, 28
102, 107, 112, 117, 124–132, 135, 137,
likelihood 2, 35, 37, 44, 49, 56, 185, 211, 221,
139, 141, 145–148, 150, 151, 153, 154,
267, 381
157–159, 161, 168–170, 173, 176,
linguistic variables 44, 46, 50, 60, 69, 72, 73,
179–183, 185, 186, 189, 192, 195–206,
208–211, 215, 221, 232, 234, 243, 247, 87, 205
255–261, 267, 270, 272, 273, 277, 293, linkages 11, 12, 139, 186, 270, 384
294, 298, 303, 312–319, 324, 325,
327–330, 334, 348, 349, 354, 360, m
369–376, 381, 384–386, 390 machine 11
investigate 11, 169, 180, 200 management 1–4, 7, 9–11, 16–25, 27–33,
investigation 2, 4, 34, 43, 68, 72, 77, 78, 81, 36–39, 41–52, 59–66, 68–74, 77–79,
87, 93, 96, 101, 103, 106, 113, 121, 142, 81–87, 89, 90, 92, 93, 102, 117,
144, 162, 168, 179, 183, 194, 196, 198, 124–136, 146–159, 161, 163–165,
202–205, 218, 221, 223, 233, 235, 239, 168–170, 172, 173, 176, 179–189, 191,
244, 245, 251, 273, 275, 332, 333, 342, 192, 194–214, 221, 225, 232, 247,
345, 350, 355, 356, 361, 388, 389 255–267, 277, 293, 296, 303, 312–324,
investigative 2, 4, 11, 34, 43, 68, 72, 77, 78, 327, 334, 348, 352, 357, 359, 360,
81, 87, 93, 96, 101, 103, 108, 113, 121, 369–381, 384, 390
142, 144, 162, 168, 179, 183, 194, 196, manual gates 1
408 Index
materials 9–13, 17, 19, 21, 26, 29, 32, 61, 93, 307, 311, 312, 341, 343, 344, 347, 348,
96, 105, 110, 114, 115, 145, 157, 159, 353, 359, 360, 364, 368, 369
192, 221, 227, 228, 233, 236, 240, 241, organisation 7, 9, 13–15, 17–19, 21–23,
245, 252, 276, 281, 283, 287, 291, 296, 25–30, 37, 41, 42, 46, 49, 59, 72, 84, 90,
300, 301, 310, 333, 339, 344, 346, 351, 92, 93, 99, 119, 148, 149, 151, 152, 155,
352, 357, 360, 366, 390 156, 163, 168, 182, 183, 185, 196, 202,
media reportage 1, 27 203, 212, 221, 227, 231, 236, 240, 245,
medicine 11 249, 285, 290, 296, 300, 305, 319, 337,
mega project 59, 83 342, 346, 351, 357, 362
methodology 11, 35, 37, 49, 50, 64, 77, 78, output 9, 12–14, 16–19, 24, 25, 29, 30, 41,
81, 85, 86, 88, 147, 153, 155, 161, 173, 43–45, 49, 55, 71, 73, 267, 324
202, 207
mixed prospects 3, 64 p
models 43, 60, 85, 125, 132, 180, 210, 211, pandemic 1, 25, 31, 32, 35–38, 41, 42
255, 261, 313, 319, 370, 376 parent companies 21, 22
modern methods of construction 27 pattern 2, 34, 51–55, 66, 68, 70, 73, 77, 78,
modern society 1, 32, 35, 36 82, 86–88, 96, 98, 100, 102, 107, 112,
modular methods 27 117, 129, 134, 146, 148, 157, 158,
multivariate statistics 34 161–165, 170, 186, 189, 192, 195, 201,
202, 230, 232, 238, 243, 247, 260, 266,
n 267, 282, 286, 293, 294, 298, 303, 317,
narratives 11, 44, 47, 60, 69, 73, 85, 125,
323, 338, 348, 349, 354, 360, 374, 380
132–136, 180, 210, 214, 255, 261–266,
people 3, 4, 47, 49, 60–65, 69, 81, 85, 124,
313, 319–323, 370, 376–380
125, 132, 136, 139, 180, 210, 211, 255,
national economy 9, 10, 29
256, 261, 267, 270, 312, 313, 319, 324,
national gross domestic product 10, 26
328, 369, 370, 376, 381, 384
natural 1, 11, 33, 38, 41, 49, 53, 68
perceptions 2–4, 7, 22, 34, 35, 38, 39, 44,
negative affect 55, 61
46–48, 50–53, 55–59, 61, 64–70, 72–74,
negative risk 57, 58, 68
77, 81–88, 91, 94, 96,–103, 107, 108,
network 11, 12, 21, 93, 183, 342, 345, 351,
112–114, 116–118, 121, 122, 124, 146,
356, 362
148, 150, 153, 157, 158, 161–166,
o 168–170, 173–177, 179, 196, 197, 199,
objective 2, 4, 9, 10, 12–14, 17, 20–25, 29–31, 201–203, 205–209, 225, 226, 229–235,
33–35, 37–39, 43, 44, 49, 52, 65–67, 238, 239, 243, 244, 247, 248, 250–252,
70–72, 74, 83–85, 91, 97, 102, 117, 162, 254, 270, 279, 280, 282, 285–289,
168, 169, 196, 199, 202, 205, 208, 209, 293–295, 298, 299, 303, 304, 306, 308,
219, 222, 225, 229, 234, 243, 279, 282, 310, 312, 319, 335, 336, 338, 341–345,
293, 303, 335, 338, 340, 348, 360 348–350, 354, 355, 360, 365, 366, 368,
Office of National Statistics 10, 26 369
open system 9, 11–14, 29, 43, 49, 71, 163 persistence 1
operational 2, 9–13, 16–18, 28, 29, 44, 52, 71, personal 1–3, 21, 23, 30, 33, 37, 39, 44, 46,
93, 100, 102, 106, 111, 116, 120, 123, 50–53, 58, 62, 64, 66–69, 73, 74, 87,
153, 165, 168, 169, 175, 179, 199, 221, 115, 126–131, 161, 163, 168, 183, 196,
227, 228, 232, 234, 237, 242, 246, 250, 202, 207, 208, 213, 257–262, 314–317,
254, 286, 288, 292, 293, 297, 302, 303, 319, 371–376
Index 409
personal perceptions 2, 39, 44, 50, 51, 53, 69, 275–277, 312–317, 319, 324, 325,
73, 74, 207 327–330, 332–334, 369–375, 381, 382,
personality traits 3, 52, 64, 69, 74, 87, 161, 384–386, 388–390
168, 196, 202 probability analysis 34, 38
perspectives 2, 7, 31, 34, 36–39, 41–43, 49, probability judgement 2, 82, 146, 150, 154,
51, 72, 77, 81, 86, 90, 100, 165, 170, 157, 159, 277, 334, 390
182, 183, 185, 232, 288, 341 probability matrix 46, 126–130, 213,
philosophy 11, 85, 168, 169, 196, 197, 205, 257–260, 314–317, 319, 371–375
206 processes 2–4, 7, 9, 11–14, 17, 20–25, 27–32,
photovoltaics 28 35, 37, 39, 41, 45, 46, 51, 52, 64, 66,
planning condition 16, 18 69–71, 74, 77, 81, 83, 84, 86, 88, 107,
policies 11, 25–27, 36, 38, 42, 45, 49, 71 117, 163, 170, 176, 180, 198, 201–203,
political 1, 10, 26, 27, 30, 31, 35, 44, 71, 93, 234, 294, 349
100, 102, 106, 111, 116, 123, 124, 165, product 9–11, 13, 14, 18, 19, 25–29, 32, 35,
167, 178, 179, 221, 231, 233, 237, 241, 36, 221
246, 250, 253, 286, 288, 297, 301, 306, product delivery 9
310, 312, 336, 338, 343, 344, 347, 353, professional 10, 14, 15, 19–22, 26, 28, 29, 41,
358, 363, 367, 369 42, 50, 53, 55, 59, 61, 69, 72, 73, 87, 92,
political functional subsystem 10, 26 94, 131, 141, 150, 153, 172, 176, 189,
political policies 26, 27 198–200, 203–209, 211, 213, 216, 219,
positive affect 57–59, 68, 393 226, 256, 262, 270, 272, 280, 282, 313,
positive heat ventilation system 28 319, 327, 330, 334, 336, 338, 370, 386
positive risk 57, 58, 68 professional services 10, 26, 29
practice 3, 4, 9, 11, 12, 17, 19–22, 24, 27, 29, project 1–4, 7, 9, 10, 13–30, 37, 39, 41–46,
30, 33, 36, 41, 43, 47–52, 55, 59–62, 48–54, 56–59, 62, 65, 66, 69, 71, 72, 74,
64–66, 68–74, 77, 81–84, 86, 87, 102, 77–79, 81–84, 86–103, 107, 110–112,
125, 127, 129–131, 133–135, 150, 153, 116–118, 120–126, 130, 131, 133, 135,
154, 157, 158, 161, 168–170, 176, 179, 140, 144–158, 161–165, 168–176,
180, 182–186, 189, 192, 195–206, 208, 179–188, 190, 191, 193–198, 201–203,
213, 214, 232, 247, 256, 259–261, 264, 205–214, 216, 218, 219, 221–223,
266, 267, 277, 293, 312, 313, 316–318, 225–238, 243, 245, 247, 248, 250–256,
322–324, 334, 348, 369, 370, 372, 374, 258, 261–263, 270, 271, 275, 276,
375, 378, 380, 381, 384, 390 279–290, 293, 294, 297, 298, 303, 304,
pre-civilization 1 306–313, 315, 317–319, 321, 324, 328,
preference reversals 3, 65, 162 332, 334–345, 348, 349, 352–354, 357,
principles 2, 9, 11, 12, 14, 17, 29, 34–36, 43, 359–362, 364–371, 374–377, 381, 384,
50, 53, 61, 63, 64, 71, 72, 77, 82, 87, 89, 388, 390
201, 204 project brief 13, 16, 18, 28
probability 2–4, 7, 32–35, 38, 44, 46, 47, 49, project management 9, 10, 17, 19, 20, 24, 30,
51, 60–65, 69, 71, 73, 82–85, 87, 39, 43, 48, 52, 71, 165, 168, 169, 197,
124–130, 135–137, 139–142, 144–147, 205
150, 154, 157, 159, 162, 179, 180, 185, project manager 2, 19, 20, 43, 53, 89, 90, 94,
186, 188, 189, 192, 195, 198–200, 204, 98, 100, 124, 135, 145, 147, 148, 150,
205, 209–211, 213, 215, 216, 218, 219, 151, 154, 155, 157, 158, 162–164, 170,
255, 257–260, 267, 268, 270–273, 171, 174, 175, 179, 181, 182, 185, 187,
410 Index
project manager (contd.) psychometric 7, 31, 33, 34, 36, 38, 39, 50, 51,
188, 190, 193, 195, 212, 226–228, 230, 56, 68, 69, 71–73, 85, 86, 88, 125, 132,
232, 255, 261, 270, 280, 282, 286, 288, 180, 210, 255, 261, 313, 319, 370, 376
312, 317, 324, 334, 338, 341, 369, 381, psychophysical scaling 34, 38
384, 390 public debate 1
project organisational level 13, 14, 21, 22, public discussions 1
29, 84, 203 public risk communication 2
public sector 25, 87, 89, 150, 151, 154, 155,
project risk 2, 24, 42, 48, 77, 81, 86, 90, 91,
158, 161, 163, 164, 170, 173, 180, 182,
93, 94, 96, 98, 107, 112, 117, 118, 121,
183, 186, 189, 195, 199
124, 125, 131, 144–148, 151, 153–155,
158, 162–164, 170, 172, 180, 181, 183,
q
186–188, 191, 194, 197, 201, 207, 212,
qualification 21
213, 218, 219, 221, 226, 230, 232, 234, qualitative 2, 32, 44, 51, 53, 60, 69, 72, 73, 85,
238, 243, 247, 248, 251, 256, 261, 262, 86, 124, 125, 132–136, 139–141, 146,
275, 276, 280, 285, 293, 294, 298, 303, 147, 150, 154, 157, 159, 161, 179, 180,
304, 308, 313, 318, 319, 332, 336, 341, 184, 185, 189–192, 198–200, 204–206,
348, 349, 354, 360, 361, 365, 370, 375, 209–211, 214, 255, 261–267, 270–272,
376, 388 277, 312, 313, 319–324, 328–330, 334,
project risk events 2, 24, 42, 77, 81, 90, 91, 369, 370, 376–381, 384–386, 390
96, 107, 112, 117, 124, 146–148, 153, qualities 29
155, 158, 162, 163, 170, 183, 201, 207, quality 16–19, 24, 25, 44, 48, 53–55, 68, 73,
219, 221, 226, 232, 234, 238, 243, 247, 77, 81, 85, 88, 93–96, 98–101, 103, 109,
280, 293, 294, 298, 303, 336, 348, 349, 113, 119, 122, 124, 163–165, 168, 175,
354, 360 179, 221, 226–233, 235, 239, 244, 248,
project success 2, 3, 42, 44, 49, 59, 62, 83, 94, 255, 280–283, 285, 287, 288, 290, 293,
169, 197, 205, 219, 226, 280, 282, 336, 295, 299, 305, 312, 336, 338, 339, 341,
338 342, 344, 345, 348, 350, 351, 353, 356,
project team members 3, 4, 21–23, 30, 56, 359, 361, 369
65, 69, 74, 83, 87, 90, 118, 121, 147, quality control inspections 19
quantity surveyor 2, 19–21, 89, 90, 94–96,
148, 151, 154, 155, 155, 158, 168, 176,
98, 100, 124, 129, 139, 147, 148, 151,
179, 196–198, 202, 203, 212, 248, 251,
154, 155, 157, 163–168, 172, 174, 175,
304, 308, 319, 361, 365
177–182, 186–189, 191, 194, 196, 198,
prospects 3, 64
199, 211, 226–230, 232, 255, 280–283,
prospect theory 3, 52, 64, 65, 69, 74, 87, 162,
288, 293, 334, 336–339, 341, 348, 360,
168, 196, 202
369, 374
psychological 3, 4, 7, 32, 33, 38, 48, 49, 56, quick approach 51–55, 68
65, 68, 71, 73, 77, 82, 87, 145, 154, 159,
173, 176, 179, 186, 189, 192, 195, 197, r
201, 203, 204, 267, 272, 277, 324, 334, rational 2, 3, 34, 39, 41, 46–51, 54, 55, 60, 68,
381, 390 69, 72, 73, 78, 79, 81, 82, 85, 87, 102,
psychological difficulties 3, 154, 173, 176, 107, 112, 125–131, 147, 150, 154, 157,
192, 195, 197, 203 169, 170, 180–186, 189, 192, 195,
psychology of perception 2, 7, 55, 163, 202 198–200, 204, 209, 210, 234, 238,
Index 411
256–261, 267, 293, 294, 298, 313–318, risk communication 1, 2, 22, 25, 30, 31,
324, 348, 354, 370–375, 381 34–36, 38, 71, 72
rational decision-making 2, 41, 50, 87, 157, risk data 4, 7, 50, 51, 60, 69, 72, 73, 85, 124,
170, 183, 204 132–134, 146, 147, 150, 154, 157–159,
rationality 46, 47, 50, 60, 68, 72, 79, 204 179, 185, 198, 201, 209, 214–216, 255,
real and objective 31, 33–35, 37 262–264, 267, 277, 312, 320–322, 324,
real and socially constructed 31, 33, 36, 38, 334, 369, 376–378, 381, 390
72 risk etymology 7, 31, 32, 201
recession 41, 42, 49 risk events 1, 2, 10, 14, 24, 30, 32, 35, 37, 39,
relationship 2, 23, 26, 36, 37, 39, 42, 52, 53, 42, 44–46, 49, 56, 59, 60, 62–65, 70,
55–57, 59, 65, 70, 74, 83, 169, 170, 182, 72–74, 77, 81, 83–85, 89–91, 93–96, 98,
196, 205 100, 102, 107, 112, 117, 120, 123, 124,
religious 1, 31, 36, 38, 71 137–139, 146–148, 150, 153, 155, 157,
remedies 1 158, 162–165, 168–170, 173, 176, 179,
repair 10 183, 196–198, 201, 203, 205, 207–209,
215, 216, 219, 221, 223, 226–228, 230,
representative heuristics 51, 62, 79, 82, 108,
232, 234, 238, 243, 247, 250, 254, 255,
176, 198, 203, 239, 295, 350
261, 267–270, 280–286, 288, 293, 294,
resource 9–13, 17, 19, 29, 48, 84, 176, 198,
298, 303, 307, 311, 312, 325–327,
203, 221
336–339, 341, 348, 349, 354, 360, 364,
review 2, 3, 7, 9, 10, 16–18, 23–26, 31, 38, 39,
368, 369, 382, 383
41, 43–46, 49–52, 54, 59–61, 64, 65, 68,
risk generation 7, 9, 86
69, 71–73, 77, 78, 81, 83, 84, 86–91, 140,
risk identification 2, 3, 31, 41, 43–45, 49, 72,
144, 146, 147, 151, 153–155, 157–159,
79, 94, 107, 112, 117, 118, 121, 124,
161, 168, 169, 179, 185, 186, 189, 192,
131, 146, 150, 153, 157, 158, 162, 164,
196, 199, 201, 202, 205, 216, 218, 271,
168–171, 173, 176, 182, 196–198, 201,
275, 312, 328, 332, 369, 384, 388, 390
203, 205, 209, 234, 238, 243, 247, 248,
rhiza 33 251, 262, 280, 294, 298, 303, 304, 308,
RIBA Plan of Work 14, 16–19, 24, 30, 53, 312, 319, 341, 348, 349, 354, 360, 361,
71 365, 369, 376
risco 33 risk interpretation 2, 9, 10, 26, 169, 196, 205
risk 1–4, 7, 9, 10, 14, 17, 19, 22, 24–39, risk management 1–4, 7, 9, 10, 19, 24, 27, 29,
41–53, 56–74, 77–79, 81–103, 107, 112, 31, 37–39, 41–52, 58–66, 68–74, 77–79,
117, 118, 120–139, 141, 144–159, 81–87, 89, 90, 92, 102, 117, 124–136,
161–173, 175–192, 194–216, 218, 219, 146–148, 150, 151, 153–155, 157–159,
221, 223, 225–235, 238, 243, 247, 248, 161, 163, 164, 168–170, 172, 173, 176,
250–252, 254–270, 272, 273, 275––277, 179–189, 191, 192, 194–214, 232, 247,
279–289, 293, 294, 298, 303, 304, 255–267, 277, 293, 303, 312–324, 327,
306–308, 310–328, 330, 332–339, 334, 348, 360, 369–381, 384, 390
341–345, 348, 349, 354, 360–362, risk management subsystems 2, 44
364–366, 368–384, 386, 388–390 risk management system 2, 10, 41, 43,
risk analysis 32–34, 37, 41, 43–46, 48, 49, 60, 45–50, 60, 61, 66, 68, 72, 74, 77, 81, 83,
61, 72, 139, 186, 205, 267, 270, 272, 85, 124–126, 150, 169, 176, 180, 183,
319, 324, 328, 381, 384 186, 197, 198, 201, 203, 205, 209, 255,
risk averse 3, 64 256, 261, 277, 312–314, 327, 369–371
412 Index
risk perception categorisation 2, 3, 7, 52, 64, science 11, 31, 32, 34, 35, 38, 51, 68, 72, 73,
65, 69, 74, 83, 86–88, 91, 96, 102, 146, 77, 86, 88, 180, 186, 189, 192, 199, 201,
148, 157, 161, 162, 164, 168, 169, 196, 202, 205, 327
197, 199, 202, 203, 205, 206, 225, 232, scientific approach 2, 36, 37
279, 293, 335, 348 Second World War 21
risk perceptions 2–4, 7, 34, 35, 38, 52, 56–58, sectors 9, 21, 22, 25, 29, 87, 89, 90, 147, 148,
64–67, 69, 70, 74, 77, 81, 83–88, 91, 94, 150, 151, 154, 155, 158, 161, 163, 164,
96–103, 112, 117, 146, 148, 153, 157, 170, 173, 180, 182, 183, 186, 189, 195,
158, 161–166, 168–170, 173, 175–177, 199, 207–211
179, 196, 197, 199, 201–203, 205–209, security risk 1
225, 226, 229–235, 238, 247, 279, 280, simulations 46, 60, 85, 125, 132–136, 180,
282, 285–289, 293, 294, 298, 303, 319, 210, 214, 255, 261–266, 313, 319–323,
335, 336, 338, 341–345, 348, 354, 370, 376–380
360–362, 364–366, 368 social construction 2, 7, 9, 10, 25, 26, 28, 35,
risk response 26, 33, 41, 43–45, 49, 66, 72, 64, 71, 77, 81, 201
124, 142, 145, 146, 158, 195, 205, 218, social construction of risk 2, 25, 26, 28, 64,
219, 273, 275, 276, 332, 388
71, 81, 201
risk review 41, 43–46, 49, 72
social hazard 1
risk variables 1
social menace 1
risky scenarios 3, 65, 69, 162
social risk communication 1
risqué 33
social risk interpretation 2
rituals 1, 31, 33, 38, 60, 71
social setup 4, 36
road 10, 12, 15, 117, 342, 345, 351, 356, 362
social structure 1, 35, 38
role 2, 9, 10, 12, 17, 20–22, 24, 30, 36, 46, 53,
socially constructed 31, 33–36, 38, 49, 72,
66, 67, 69, 71, 83, 84, 87, 90–92, 95–98,
86, 88
100, 102, 117, 124, 129, 146, 148–158,
socially mediated 31, 33–35, 38
161–165, 168–176, 179–182, 184,
society 1, 4, 31, 32, 34–36, 38, 162
186–190, 192–199, 202–208, 211, 212,
socio-cultural functional subsystem 10, 27
225, 227–230, 232, 247, 255, 261, 267,
sociology 38
279, 281–283, 286, 288, 293, 303, 312,
317, 334, 335, 337–339, 341, 348, 360, specialist installations 10
369, 374, 390 specialist roles 2, 9, 10, 17, 20–22, 24, 30, 46,
Romans 11 66, 67, 71, 83, 84, 90–92, 96–98, 100,
102, 117, 124, 129, 146, 148–158,
s 161–165, 168–176, 179, 181, 182, 184,
safety 1, 16, 22, 28, 61, 90, 93, 95, 96, 100, 186–190, 192–199, 202–208, 225,
106, 111, 116, 120, 143, 150, 216, 221, 227–230, 232, 247, 255, 261, 267, 279,
228, 232, 234, 237, 242, 243, 246, 250, 282, 286, 288, 293, 303, 312, 317, 334,
253, 274, 283, 297, 330, 333, 339, 343, 335, 338, 341, 348, 360, 369, 374
344, 347, 353, 359, 364, 367, 386 specification 19, 45, 89, 93, 114, 221, 245,
safety guidance 1 281, 296, 300, 356
scenario 56, 59, 65, 82, 126–130, 213, 248, stage 9, 11, 13, 14, 16, 17, 21–24, 28, 30, 39,
251, 257–260, 262, 314–317, 371–375 41, 43–46, 49, 53, 59, 71, 81, 85, 86, 88,
schools 10, 12, 29, 35, 59, 63, 343, 344, 347, 90, 93, 96, 97, 102, 109, 112, 169, 209,
353, 359, 364, 368 219, 222, 227, 229, 234, 238, 239, 281,
Index 413
282, 293, 295, 298, 337, 339, 340, 348, 267, 272, 277, 279, 280, 282, 293,
350, 352, 354 312–314, 324, 327, 334–336, 338, 342,
stakeholder 4, 13, 28, 29, 74, 93, 162 345, 348, 351, 356, 362, 369–371, 381,
statistical 4, 7, 51, 60–62, 64, 69, 73, 78, 84, 390
85, 87, 88, 124, 125, 132, 137, 139, system theory 9, 17, 72
145–147, 150, 154, 157, 159, 161, 179, systematic errors 51, 54, 63, 69, 73, 82, 185,
180, 185–187, 198, 199, 204, 205, 209, 195, 199, 205, 324, 327
210, 215, 255, 261, 267, 268, 270, 276, systems components 4, 13, 17
277, 312, 313, 319, 325, 327, 333, 334, systems decomposition 4, 9, 14, 15, 17, 20,
369, 370, 376, 381, 382, 384, 390 23, 29, 41, 43, 49, 52, 65, 70, 71, 74, 83,
statistical generalisation 78, 88, 161 86
statutory 2, 11, 13, 16, 18, 28, 32, 44, 93, 94, systems differentiation 2, 4, 10, 17, 19, 20,
98, 100, 101, 105, 111, 115, 123, 124, 24, 29, 30, 66, 67, 77, 83, 84, 86, 87, 91,
163, 165, 168, 175, 179, 226–228, 102, 146, 148, 153, 162, 163, 201–203,
230–233, 237, 241, 246, 249, 253, 255, 225, 232, 279, 293
280–283, 286, 288, 291, 297, 301, 306, systems objectives 9, 13, 29
310, 312, 336–339, 341, 343, 344, 347, systems of thinking 3, 7, 51, 52, 68, 72, 73,
348, 352, 358, 363, 367, 369 86, 168, 180, 199, 201, 272
statutory regulations 28 systems thinking 2, 7, 9, 11, 14, 28, 29, 36,
stratified vertical system 1, 31 37, 39, 41, 43, 61, 77, 86, 201, 202, 204
structure 11, 12, 20, 22, 23, 25, 28–30, 35, 49,
63, 74, 82, 87, 93, 101, 102, 104, 109, t
114, 122, 146, 168, 180, 196, 201, 202, taboos 1, 27, 31, 38, 60, 71
344 task 13, 19, 47, 51–54, 58, 68, 69, 73
structured risk identification 2 taxation 28, 115, 221
subgroups 4, 22, 24, 30, 65, 94, 162, 202, 203, technical manager 2, 89, 90, 94, 98, 100, 124,
226, 280, 282, 336, 338, 341 134, 147, 148, 150, 154, 155, 157, 158,
subject 11, 12, 29, 37, 39, 56, 60, 69, 85, 102, 162, 163, 171, 174, 175, 179–181, 187,
117, 161, 168, 169, 196, 209, 234, 243, 188, 190, 193, 226, 230, 232, 243, 255,
293, 303, 348, 360 270, 336, 341, 369, 374, 390
subjective analysis 38 techniques 3, 4, 9, 27, 29, 38, 44, 46, 48, 50,
subjectively biased 31, 33, 34, 38 57, 60, 69, 71–73, 84, 86, 87, 126–131,
subjectivity 2, 77, 85, 102, 112, 169, 173, 174, 134, 135, 147, 150, 154, 157, 158, 182,
201, 203, 209, 234, 238, 293, 294, 348, 183, 185, 198, 200, 201, 204, 213,
354 257–261, 267, 272, 277, 313–318, 324,
subjectivity of risk perceptions 2, 77, 201 370–375, 381
superstition 1, 27, 31, 38, 60, 71 technological 10, 11, 27, 36
surrounding 11 technological functional subsystem 10, 27
sustainability consultant 19, 20, 22 tender selection question 28, 43
system 1–4, 9–31, 33–39, 41, 43–53, 56, 58, theoretical 2–4, 7, 9, 10, 17, 26, 30, 31, 33,
60–74, 77–79, 81–87, 91–94, 97, 102, 37–39, 41, 49–53, 55, 56, 59, 61, 64–66,
124–126, 129, 134, 135, 146–154, 68, 69, 71–74, 77, 78, 81–83, 86–89, 91,
156–158, 161–163, 168–170, 176, 179, 96, 97, 102, 107, 112, 117, 124, 127,
180, 182, 183, 185, 186, 189, 195–205, 132, 134–137, 139, 141, 145–148, 150,
207–210, 225, 229, 232, 255, 256, 261, 153, 154, 157–159, 161–164, 168–170,
414 Index