Risk in structural engineering

October 2013
Membership of the Task Group
C J Bolton BSc CEng FIStructE MICE (Sellafield Ltd) Chairman
R A Davis MSc CEng FIStructE (TPS)
J K Kenward BEng(Tech) CEng FIStructE MICE MCIHT (Hyder Consulting Limited)
J Lane* MSc CEng FICE (RSSB)
Dr A P Mann FREng BSc(Eng) PhD CEng FIStructE MICE (Jacobs)
B S Neale CEng FIStructE FICE Hon FIDE (Consultant and Hazards Forum)
D A B Thomas BSc(Eng) MSc AKC CEng FICE CFIOSH (The heightec Group Ltd)

Corresponding members
D H Bardsley BSc ACGI CEng FICE (Consultant)
Dr W G Corley** PhD CEng FIStructE (CTL Group)
A M Cormie BSc CEng FIStructE FICE FIES MWeldI (J&D Pierce (Contracts) Ltd)
K K Kwan CEng FIStructE MICE FHKIE (Arup)

Secretary to the Task Group

Dr J D Littler PhD (The Institution of Structural Engineers) (until June 2011)
B Chan BSc(Hons) AMIMechE (The Institution of Structural Engineers) (from June 2011)
A Rahman MEng(Hons) (The Institution of Structural Engineers) (from January 2013)
* representing ICE
** deceased March 2013

Acknowledgements
Figure 3.1: Chris Bolton
Figure 4.1: Contains public sector information published by the Health and Safety
Executive and licensed under the Open Government Licence v1.0
Figure 5.1: TPS
Figure 7.1: Mandy Reynolds (ss Great Britain Trust)
Figure 7.2: Peter Smyly
Figure 8.1: Guy Gorton
Figure 8.2: Chris N Illingworth
Figure 9.1: Minnesota Department of Transport
Figure A.1: Nicholas Smale
Figure A.2: Lee L Lowery, Jr., PE PhD
Figure A.3: CA Group
Figure A.4: U.S. Department of Defense

Published by the Institution of Structural Engineers

International HQ, 11 Upper Belgrave Street, London SW1X 8BH
Telephone: þ44(0)20 7235 4535 Fax: þ44(0)20 7235 4294
Email: [email protected], Website: www.istructe.org
First published 2013
ISBN 978-1-906335-08-3

#2013 The Institution of Structural Engineers

The Institution of Structural Engineers and those individuals who contributed to this Report have
endeavoured to ensure the accuracy of its contents. However, the guidance and recommendations
given in the Report should always be reviewed by those using the Report in the light of the facts of
their particular case and specialist advice obtained as necessary. No liability for negligence or
otherwise in relation to this Report and its contents is accepted by the Institution, the members of
the Task Group, their servants or agents. Any person using this Report should pay particular
attention to the provisions of this Condition.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any
form or by any means without prior permission of the Institution of Structural Engineers, who may
be contacted at 11 Upper Belgrave Street, London SW1X 8BH.
Glossary and abbreviations

These definitions are provided to explain how the terms listed are used in this Report, and are not necessarily
as used in other documents. Note that ‘risk’ and ‘hazard’ are defined in the singular; in practice there will usually
be many hazards and even more risks.

Term Definition
Hazard The potential for harm arising from an intrinsic property or disposition of something to cause detriment.
Risk A combination of the severity of the harm resulting from a hazard and the probability that the harm is
realised.
Constructor The person or organisation responsible for execution of the work. Sometimes referred to as the
contractor.
Designer Any person or organisation designing permanent or temporary works.
Elimination Avoidance of risk by changing the design (or, possibly, the construction method) so that the hazard
ceases to exist.
Structural engineer Any qualified structural engineer, such as a Chartered, Associate-Member or Technician Member of the
Institution of Structural Engineers, or the equivalent in another body worldwide, whether working on
behalf of a designer, constructor or client.
Harm Any unwanted consequence, including death, injury, damage to health and serious economic detriment.
Minimisation Measures that reduce the probability that the risk occurs.
Mitigation Measures that reduce the severity of the harm, given that the risk occurs.
Residual risk Residual risk is the level of risk remaining when the current risk control measures and their degree of
effectiveness are taken into account.
Risk assessment A document recording the conclusions of a risk assessment.
document
Risk assessment A process that identifies hazards, evaluates risks and informs its users about risks, their possible
elimination, the controls that might be put in place to reduce them and the residual risks that might
remain.
Risk management The whole process of identifying risks and responding to them (not just the control measures).
Safety culture (of an The combination of individual and group values, attitudes, perceptions, competencies, and patterns of
organisation) behaviour that determine the commitment to, and the style and proficiency of, an organisation’s health
and safety management.

Abbreviation Definition
ACoP (UK) Approved Code of Practice
ALARP As Low as Reasonably Practicable
ERIC Eliminate, Reduce, Inform, Control
CDM (UK) Construction (Design and Management) Regulations
HAC High Alumina Cement
HAZOP HAZard and OPerability review
HSE (UK) Health and Safety Executive
SFAIRP So Far As Is Reasonably Practicable (may also be written SFARP)
QA Quality Assurance
QRA Quantitative Risk Assessment

The Institution of Structural Engineers Risk in structural engineering v

Contents

Glossary and abbreviations v 5.4.3 Documenting the risk assessment 21

5.4.4 What to do and not to do in risk
Foreword vi assessment 23
5.5 HAZOP 23
1 Scope and objectives 1 5.6 Codes of practice 23
1.1 Introduction 1 5.7 Risk management framework 23
1.2 Types of risk 1 5.8 The importance of competence 24
1.3 Intended readers 1 5.9 QA and change control 24
1.4 Relevance to clients 1 5.10 Independent review 24
1.5 Summary of objectives 1 5.11 Conclusions and recommendations 24
1.6 References 2 5.12 References 25

2 Hazards and risks 3 6 Statistical and probabilistic methods 26

2.1 Introduction 3 6.1 Introduction 26
2.2 The meaning of risk and hazard 3 6.2 Background 26
2.3 Structural engineering hazards 3 6.3 Quantifying probability 26
2.4 Structural engineering risk 3 6.4 Safety factors 27
2.5 Summary 5 6.5 Low probability events 27
2.6 References 5 6.6 Application 28
6.7 Assessment of existing structures 28
3 Principles of risk management 6 6.8 Conclusions and recommendations 29
3.1 Introduction 6 6.9 References 29
3.2 The importance of risk management 6 6.10 Bibliography 30
3.3 How big is the risk? 6
3.4 Competence 7 7 Risk in design 31
3.5 Communication 7 7.1 Introduction 31
3.6 What is an acceptable level of risk? 7 7.1.1 Overview 31
3.7 Proportionality and ALARP 8 7.1.2 What does the design stage include? 31
3.8 Risk management hierarchy 8 7.1.3 The consequences of design stage errors 31
3.9 Risk aversion 8 7.1.4 What can go wrong? 31
3.10 Resilience 9 7.1.5 Managing uncertainty 31
3.11 Ownership and control of risks 9 7.1.6 Project risk awareness 32
3.12 Societal risks 9 7.2 Managing the design process 32
3.13 Human failure and accidents 9 7.3 Clarity of responsibility 33
3.14 Safety culture 10 7.4 Design changes 33
3.15 Prescription versus engineering judgement 10 7.5 Clarity of design requirements 33
3.16 The safety case 11 7.6 Designing robust structures 33
3.17 Conclusions and recommendations 11 7.6.1 Principles of robustness 33
3.18 References 11 7.6.2 Designing for accidental loads 34
3.19 Bibliography 12 7.7 Designing for construction 34
7.7.1 Buildability 34
4 Legal background 13 7.7.2 Designing out construction hazards 35
4.1 Introduction 13 7.7.3 Communication of risk 36
4.2 Law as it affects structural engineering 13 7.7.4 Designing for unfamiliar environments 36
4.3 Acts, regulations, guidance and ACoPs 13 7.8 Design for the whole building life cycle 36
4.4 Reasonably practicable 13 7.9 Designing for future demolition 36
4.5 Practicable 14 7.10 Procurement and planning 37
4.6 Burden of proof 14 7.11 Conclusions and recommendations 37
4.7 Liability under civil law and duty of care 14 7.12 References 37
4.8 Law enforcement 15
4.9 Designer’s role 15 8 Risk management during construction 39
4.10 Building Regulations 15 8.1 Introduction 39
4.11 Europe 15 8.2 Causes of incidents 39
4.12 Hong Kong 16 8.3 Resourcing and planning 39
4.13 United States 16 8.4 Competence, management and welfare 40
4.14 Summary and conclusions 16 8.5 Communication 40
4.15 References 17 8.6 Loading conditions 40
4.16 Bibliography 17 8.7 Sequence of construction 41
8.8 Temporary works 42
5 How to manage risk 19 8.9 Protective equipment 42
5.1 Introduction 19 8.10 Access and work areas including working at
5.2 First identify the hazards 19 height 42
5.3 Apply the risk management hierarchy 19 8.11 Lifting 42
5.4 How to do ‘risk assessment’ 20 8.12 Conclusions and recommendations 43
5.4.1 General 20 8.13 References 43
5.4.2 Process for risk management 21 8.14 Bibliography 44

The Institution of Structural Engineers Risk in structural engineering iii

9 Risk management during the life of a structure 45
9.1 Introduction 45
9.2 The life cycle of a structure 45
9.3 Risks during the life of a structure 45
9.4 Risk management strategy 46
9.5 Maintenance regime 46
9.6 Appraisal and assessment 47
9.7 Conclusions and recommendations 47
9.8 References 48
9.9 Bibliography 49

10 Risk management for demolition and

refurbishment 50
10.1 Introduction 50
10.2 The potential for unplanned events 50
10.3 The need to know the existing structure 50
10.4 Structural refurbishment 50
10.4.1 Overview 50
10.4.2 Structural stability 51
10.4.3 Fire precautions 51
10.4.4 Underpinning works 51
10.5 Managing (deliberate) structural instability 51
10.6 Reducing uncertainty 51
10.7 Guidance 51
10.8 Conclusions and recommendations 52
10.9 References 52
10.10 Bibliography 52

Appendix A Case studies 53

A.1 Introduction 53
A.2 Uncertainty in loading 53
A.3 Extension of technology to an invalid extent 53
A.4 Fatigue loading 54
A.5 Uncertainty in extreme loading 54
A.6 Failure to understand materials 54
A.7 Failure to identify the hazard 54
A.8 Errors in dynamics 55
A.9 Errors in stability 55
A.10 Errors in design or detailing 55
A.11 Deterioration and lack of maintenance 55
A.12 Identifying significant risks 56
A.13 Demolition 56
A.14 Human factors 56
A.15 Design change 56
A.16 Temporary works and construction failures 56
A.17 Inadequate procedures 57
A.18 Systems failures 57
A.19 Robustness 57
A.20 Mobile structures 58
A.21 Failure to learn from previous incidents 58
A.22 Safety culture 58
A.23 Competence and quality 59
A.24 Failure to understand the structure 59
A.25 Novel design 59
A.26 References 60

iv The Institution of Structural Engineers Risk in structural engineering

1 Scope and objectives

1.1 Introduction This Report has been compiled by a group of

practising engineers and defines the Institution of
Structural Engineers’ view of good practice in risk
This chapter of the Report defines the scope. This management. As legal requirements vary between
includes what kinds of risk are covered, who the countries the Report does not, in general, define how
intended readers are and what, in broad terms, the to comply with the law. The exceptions to this are the
Report seeks to achieve. chapter on legal requirements and where reference to
laws is used to illustrate a principle. So far as the
authors are aware, none of the guidance conflicts
with any legal requirements.
1.2 Types of risk
The Report is divided into chapters but important
concepts are illustrated throughout. Anyone
This Report covers all risks which fall to structural concerned with risk in structural engineering, at any
engineers to manage. Primarily, this means risks to stage in the life of a structure, is encouraged to read
the health or safety of workers and the public, and the whole Report.
serious environmental damage, where there are both
legal and moral obligations to keep risks low.
Significant cost and programme risk arising from
structural engineering decisions is also in the scope. 1.4 Relevance to clients
Business risk in engineering projects is not unique to
structural engineering and is covered elsewhere1.1, 1.2,
so is not included. While structural engineers, as designers or
constructors, should have the experience and
Communication and co-ordination is an essential part competence to understand and manage risks,
of risk management. The structural engineer is in a clients also have a significant role. They are
good position to take a lead role on suitable projects responsible for providing sufficient information, time
and should develop the competences required to do and resources to ensure that risks can be properly
this. managed, and they often have essential knowledge
on the use and lifetime management of the
The scope is not limited to design of new works; structure. The engineer should ensure that the
risks will occur throughout the life cycle of a building client understands these responsibilities. Clients
or structure; from concept through detailed design, should take care as to whom they allocate risk, as
construction, commissioning, operation, some risks are best managed by the client.
maintenance, dismantling, demolition and ultimately Engineers are encouraged to point out to clients
to disposal. the value of sound risk management, underpinned
by a culture of safe design, and to support them in
achieving this.

1.3 Intended readers

1.5 Summary of objectives
The objective of this Report is to help structural
engineers in all parts of the industry to identify risks
and deal with them. ‘Structural engineers’ includes The Report promotes a proportionate and open
not only designers and consultants but also approach to risk assessment and management as a
engineers engaged in site work, existing structures process for preventing harm, not for producing
and demolition. An understanding of the whole documents. It aims to put more meaning behind the
picture is important for engineers in all sectors and words ‘risk in structural engineering’ by describing
will assist exchange of useful information to reduce the types of risks that may affect structural
risk. The Report is relevant to all structural engineers, engineering work and by providing a decision making
from students and graduates learning how to framework for risk, taking account of legal principles,
manage risks to experienced engineers and available guidance and research. As set out by the
technicians involved with any scale of project from UK’s Engineering Council1.3, good risk management
domestic improvements to unusual or high risk requires engineers to make judgements, avoid risk
projects. aversion and hence facilitate innovation. The Report
describes the legal background, in the UK and
The Report is mainly based on UK experience but the elsewhere, and provides examples of risks that were
principles of good practice are applicable worldwide. managed badly and that resulted in serious accidents.
Local laws, cultures and economic circumstances Tools, techniques and selected references to assist
may require a different approach in detail. Although engineers are described, although once the principles
the recommendations deal with structural are understood, managing risk is mostly a matter of
engineering, examples of risk issues from related attitude, culture and common sense. The Report
industries are used to illustrate common themes and cannot say everything about risk in structural
to show the potential to apply learning from any engineering, but it allows the reader to make further
source. studies.

The Institution of Structural Engineers Risk in structural engineering 1

1.6 Scope and objectives

1.6 References

1.1 Actuarial Profession and Institution of Civil Engineers.

Risk analysis and management for projects [RAMP].
2nd ed. London: Thomas Telford, 2005

1.2 Actuarial Profession and Institution of Civil Engineers.

Handling uncertainty – the key to truly effective
enterprise risk management. Available at:
http://www.actuaries.org.uk/sites/all/files/documents/
pdf/ermguidehandlinguncertainty_0.pdf
[Accessed: 18 February 2013]

1.3 Engineering Council. Guidance on risk for the

engineering profession. Available at: http://www.engc.
org.uk/risk [Accessed: 18 February 2013]

2 The Institution of Structural Engineers Risk in structural engineering

2 Hazards and risks

2.1 Introduction damage to health, pollution, delay to the project, or

the completed structure being unserviceable or
unstable.
This chapter introduces the topics of hazard and risk,
together with their relevance to structural engineering. If a hazard can be removed entirely, the risk will
It discusses the type of hazards and risks that disappear. It is never too early to consider elimination
structural engineers should address. of hazards; if a hazard is inherent in the concept
design, it is often not possible to remove it, only to
manage the risk.

2.2 The meaning of risk and hazard Some hazards are outside the control of engineers or
cannot be eliminated. These include frequent events
like ‘wind loading’, rare events such as ‘earthquake’
Risk is an integral part of everyday life and one of the or ‘out of control vehicle’ and less well defined
most widely considered topics in modern society. hazards such as ‘terrorists’. A higher probability of
Almost every activity includes some risk. This may be failure against risks from less frequent hazards may
obvious or unrevealed, well or poorly understood and be considered tolerable, but good engineering can
may then be accepted or ignored. Written ‘risk still reduce that risk.
assessments’ are now produced for many activities
but these often miss the point, which is not to Within the construction process itself there are well
produce documents but to remove hazards and to known hazards such as ‘work at height’ and hazards
understand, communicate and manage the remaining to health such as ‘asbestos’ or ‘wet concrete’ (risk of
risks. dermatitis). There are accepted ways to deal with
these; some can be eliminated in design, others by
The practice of structural engineering acknowledges choice of construction method and the risk from the
risk as an inescapable reality. If the structure is not remainder can be minimised. Other hazards, such as
designed or executed properly it may fail with instability of the permanent or temporary works, are
significant consequences, including a liability on those specific to the project and require individual
at fault. consideration. Particularly on complex projects, these
may not be obvious unless a structured approach is
In everyday language, ‘risk’ refers to danger, peril, taken to identifying them. A list of common hazards
and exposure to loss, injury or destruction. In that structural engineers might influence is included in
engineering terms, it is useful to distinguish hazard Figure 2.1. Useful additional information is given in
from risk. In this Report, a ‘hazard’ is something with CIRIA reports C6622.1 and C6632.2; although written
the potential to cause harm. For example, fire is a to address UK legislation these are applicable
hazard; death, injury and loss of property as a result elsewhere.
of fire are risks.
In some cases, hazards that led to serious structural
Risk is a combination of the likelihood of the harmful failures may not, and in some cases could not, have
event occurring and the consequences if it did occur. been anticipated by the designer. To take two
Depending on the type of risk, the likelihood might be unconnected examples, the causes of the failure of
expressed as a probability (e.g. 2%, 1 in 1000) or in the terminal building at Charles de Gaulle Airport2.3
words (e.g. barely credible, likely, unlikely, possible). and the collapse of the World Trade Center2.4 may not
Similarly the consequences could be identified by have appeared on any list of hazards.
descriptions such as injury, fatality, cost or delay.
These might then be quantified, if information is
available, using details such as ‘loss of injured
person’s sight’, ‘cost of £2 million’ or ‘up to six 2.4 Structural engineering risk
weeks delay’. As a method of combination,
risk ¼ likelihood  consequence captures the broad
concept, but it is often not possible to be Risk is not a new subject for structural engineers.
mathematically precise. While 10%  £2 million is Ways to manage the risk of structural failure have
clear, it becomes meaningless to multiply descriptive been evolved over hundreds of years. The profession
terms, such as ‘unlikely’  ‘up to 6 weeks delay’. In has developed codes of practice and ways of
addition, for high consequence risks, the working that will, most of the time, ensure that
consequence is often given greater weight than the structures are adequate for the loads they are called
likelihood. on to resist. With knowledge of the operating
conditions, loads, environment, etc., it is possible to
design structures with high confidence of satisfactory
performance. Each structure is unique, however, with
2.3 Structural engineering hazards the result that structural engineering is characterised
by the production of ‘prototypes’, none of which will
be tested until put into service, and will probably
A hazard is anything that may cause harm. In the never be tested against accidental loads unless an
context of structural engineering, that means anything accident happens. In practice, knowledge of the
that can go wrong. Examples include death, injury, operating conditions is usually imprecise. Although

The Institution of Structural Engineers Risk in structural engineering 3

2.4 Hazards and risks

(a) At any time (b) In design and construction (c) In use, between uses, after use
At any time: In design: In use:
To structure and people To structure and people To people in building
– Inappropriate concept (risks that arise later – Misunderstandings in design (including – Stairs, floor finishes, glazing
cannot be managed) thoughtless use of computers)
To structure and people
– Inadequate resources, short-cuts, – Omission of load cases; neglect of dynamic
inexperienced/incompetent staff effects; thermal movements; deflection; – Inadequate maintenance
– Poor communication, co-ordination inadequate consideration of structure-soil – Change of use
– Inappropriate procurement interaction – Inadequate access
– Pressure of time and money – Neglect of constructability, maintainability
and demolition In maintenance:
– Inappropriate tolerances
– Novel materials and design concepts To people doing maintenance
– Defective, unsuitable, undersized or badly
– Access, confined spaces
specified materials
– Hot materials, toxic materials
– Foundations: settlement, chemical attack
– Falls from height, fragile roofs
– Unexpected ground conditions (soil strength,
contamination, chemical effects)
– Unexpected or accidental loads: overload,
misuse, weather, terrorism, explosion,
impact, temperature In extension, refurbishment and repair:
– Natural events, e.g. floods
– Instability – Misunderstanding the original structure
– Unidentified critical element(s) – Faults in the original structure
– Lack of redundancy or other robustness – Earlier inappropriate modifications
– Fire In construction: – Fire-fighting and emergency services
– Corrosion and ageing, dynamic effects, To people doing construction
progressive/disproportionate collapse
– Risks to, or from, adjacent buildings, – Falling materials, excavations, falls from
structures and other facilities height
– Manual handling, toxic materials, vibration, In assessment:
noise – Incorrect assumptions (materials,
– Vehicular movements structural form, loads)
– Confined spaces – Inadequate inspection
– Putting people close to hazards (e.g. site – Structural behaviour
office below bridge construction) – Incorrect analysis
To structure and people
– Failure of critical elements
– Instability of part completed structure
In demolition:
– Temporary works failure due to instability,
capacity, foundations or concept – Misunderstanding structure
– Defects in structure
– Inappropriate approach
– Premature collapse, flying debris
– High risk elements; cantilevers, flat slabs,
prestressed structures, retaining structures

Note
This list is NOT exclusive and is provided as an example of hazard identification

Figure 2.1 Some hazards that may be encountered in structural engineering

much has been learned over the years, not every machinery is used. There are few work environments
technique is tried and tested, with new methods and with as many different risks.
products being constantly introduced against a
background of pressure to reduce costs and deliver The type of risks to be considered at each stage of
faster. the design and construction process will differ. It is
important to recognise the general impossibility, for
The emphasis on managing risk to people working both technical and cost reasons, of being able to
on construction is more recent. Construction eliminate all hazards or reduce risk to zero. Hence,
generally involves several different companies, through an experience based decision making
requiring communication across the interfaces. The process, engineers may choose to ‘accept’ a certain
site changes as construction proceeds, resulting in degree of risk, but that has to be done with proper
risk changing daily. It is exposed to weather and understanding, logic and justification.
controlling access can be difficult. The structure itself
and the materials used to build it are heavy, meaning While the risks of bad weather and poor ground
that substantial forces are involved and powerful conditions may appear, initially, to be outside the

4 The Institution of Structural Engineers Risk in structural engineering

 Hazards and risks 2.5

control of the structural engineer, they are 2.4 National Commission on Terrorist Attacks on the United
nevertheless risks which can be reduced if they are States. The 9/11 Commission Report. Available at:
given proper consideration. Similarly, structural http://govinfo.library.unt.edu/911/report/911Report.pdf
engineers can make a beneficial input in reducing the [Accessed: 18 February 2013]
risks in procurement, such as a less than competent
constructor or an inappropriate contract form for the 2.5 ‘Almuñécar: seeds of a falsework failure’. New Civil
type of structure. Engineer, 24 November 2005, p12

Structural failures can also occur during construction

and there have been a number of dramatic examples
such as the falsework failure of the Almuñécar
bridge2.5 in Spain (2005). Failures rarely have a single
cause, but a significant contributor to many of these
failures was poor communication between temporary
and permanent works designers or failure of designers
to understand the loads applied during construction.
Some of these failures occurred on large projects
involving major companies. This highlights the
existence of risk and the fact that no design team is
immune.

Failures have occurred throughout the history of

structural engineering, often as a result of hazards
that were not identified or risks that were
underestimated in terms of probability or
consequence. Each generation of engineers is faced
with new challenges. Unfortunately, many of the old
challenges remain and, in some cases, the same
mistakes are made. There are many useful lessons in
what could go wrong and how it might be prevented,
but ‘corporate memory’ is hard to retain. The starting
point for promoting knowledge of ‘risk’ is to study
incidents and their causes. Appendix A includes a
number of case studies.

2.5 Summary

The hazards and risks in structural engineering can

be considered in terms of structural stability, health
and safety of workers and commercial
consequences. Construction is at the centre of
structural engineering and involves multiple interfaces,
a constantly changing site environment and handling
of heavy materials. This results in many hazards and
risks which can be severe in both probability and
consequence.

Structural engineers have a legal and professional

duty to eliminate hazards and minimise the probability
and consequences of any remaining risks. Some are
easy to identify and manage; others require detailed
consideration and a structured approach. Chapter 3
will discuss the broad principles and issues in this
task, Chapter 4 will define the legal constraints and
Chapter 5 will give specific recommendations.

2.6 References

2.1 Gilbertson, A. CDM 2007 – Construction work sector

guidance for designers. C662. London: CIRIA, 2007

2.2 Gilbertson, A. CDM 2007 – Workplace ‘‘in-use’’

guidance for designers. C663. London: CIRIA, 2007

2.3 Wood, J.G.M. ‘Paris airport terminal collapse: lessons

for the future’. The Structural Engineer, 83(5), 1 March
2005, pp13-14

The Institution of Structural Engineers Risk in structural engineering 5

3 Principles of risk management

3.1 Introduction at risk minimisation may be best applied, it is

necessary to understand how these factors affect the
size of each risk, so that they can be given
This chapter discusses structural engineering risk in consideration in proportion to their seriousness.
general terms; how and why it may arise and some of
the principles that influence risk management. Neither structural engineering nor risk management is
Specific guidance on how to apply risk management an exact science. While the probability of a particular
tools is given in Chapter 5. load, such as wind, can be estimated, it is generally
difficult to quantify structural risk in numerical terms.
Although many techniques exist to attempt this,
some of which are discussed in Chapter 6, their main
3.2 The importance of risk management use is to make comparisons between risks.
Assessment of residual risk for different options can
help to select the best approach. Most risks can be
In most countries, the accident rate in construction is
minimised by following good practice and using
well above the average for industry in general. Even
engineering judgement which, in turn, requires the
without accidents, many common construction
kind of background information this Report provides.
practices can affect workers’ health. Unforeseen
events can cause projects to run seriously late or
Even within the profession, there will be different
over budget. Frequently, the lack of foresight occurs
attitudes to risk; engineers are required to produce
during the structural design phase or in the detailed
safe structures against a background of uncertainty in
planning for construction.
loading, using materials that have variable properties,
all supported off uncertain ground conditions.
Every risk is a learning opportunity. Only a small
percentage of structures fail in harmful ways but
To understand a risk it is necessary, among other
minor mishaps are more common. The evidence from
things, to have a realistic or bounding estimate of the
individual tragedies and ‘near misses’ is crucial to
probability of the initiating event. Perceptions of risk
education in assessing risks and producing safer
vary, and are not always an accurate pointer to areas
designs, particularly for low probability, high
for risk reduction. Some risks may have high probability
consequence events.
but limited consequences, while others may have low
probability with serious consequences. If there are risks
All parties have an obligation to assess and minimise
with high probability and high consequence they are
risks that might cause personal harm and, in some
unlikely to be tolerable; the hazard should be eliminated
jurisdictions, specific roles and records are required.
or the project would not be viable.
Many engineers will also perceive a profound self
interest in minimising risks that might damage their
Engineers often work with unknowns, and have
professional lives. There is also good economic
developed methods to manage this. While the actual
justification for taking the trouble to manage risks
strength of a given section of concrete is unknown,
properly. Failures and accidents can be very expensive.
as is the highest wind speed next winter, statistical
The cost of structural failure invariably exceeds the
methods and factors of safety enable engineers to
costs of preventing the incident by a significant margin.
have appropriate confidence in their designs. Wind
In addition to the human cost, a single fatality can
speed and concrete strength may be defined as
result in millions of pounds in direct and indirect costs,
‘known unknowns’, because the events they relate to
such as stopping work, internal inquiries and contract
have been identified.
penalties, with the possibility of prosecution.
There are also ‘unknown unknowns’, where even the
The true cost of accidents and ill-health resulting from
possibility has not occurred to the engineers involved.
work is often underestimated. Research in the UK
If the designers of the World Trade Center had
has reported that various employers estimated their
anticipated that terrorists might deliberately crash a
costs at £30000 to £2 million per year3.1. Large
large fully fuelled aircraft into their structure, they
organisations tend to have a better understanding, due
might have been able to estimate the range of
to the number of accidents; in a small company,
possible impact parameters, or ‘known unknowns’.
accidents are infrequent and the costs are not
As it was, the event was outside anyone’s
appreciated. The cost to employers of workplace
imagination at the time of design3.3. Completely
injuries, work-related ill-health and accidental damage
unknown events are sometimes known as ‘black swan’
events in the UK has been estimated to be between
events3.4. Until the first black swans were discovered in
£3.5 billion and £7.3 billion a year3.2. The economic
Australia, ornithologists never considered the
cost and disruption to society from infrastructure
possibility that swans could be anything other than
failures such as loss of a bridge is almost incalculable.
white. A further example is the failure of the Nishinomiya
Harbour Bridge during the ‘Kobe’ Earthquake
(1995)3.5. The made ground slumped towards the
3.3 How big is the risk? harbour, taking the bridge foundation with it and
causing the side span to lose its bearing (Figure 3.1); it
appears that this failure mode was not anticipated by
The combination of the type of harm, its severity and the designer. Robustness and sensitivity studies are the
its probability defines the risk. To decide where effort best approach to managing such unknown risks.

6 The Institution of Structural Engineers Risk in structural engineering

 Principles of risk management 3.4

Figure 3.1 Nishinomiya Harbour Bridge

3.4 Competence and safety matters, which in many cases means they
should be the best people to co-ordinate the design
to minimise risk.
Competence is crucial to risk management; decisions
should only be made by people who have Communication is not only necessary from engineer
appropriate skills, knowledge and experience. to engineer, but between engineers and the public. It
Sometimes, this requires involvement of more than is very easy to find that public opinion is opposed to
one person. A group of experienced people, using a a development because the risks (both from building
brainstorming approach, will often identify risks and it and from not building it) have not been fully
ways to reduce them which would not occur to an explained. There was controversy in 2012 when a
individual working alone. number of Italian earthquake scientists were
convicted for apparently giving inappropriate
Ensuring adequate competence and resources is a reassurance that a severe earthquake in L’Aquila3.8
fundamental part of procurement for any contract, be was not imminent. The error was said to be not in their
it for design, construction, maintenance or demolition. scientific predictions, but in the way these were
Passing risk along the supply chain may appear to expressed to the public.
save money, but it will not do so if the risk becomes
the responsibility of somebody who is unable to
manage it.
3.6 What is an acceptable level of risk?

3.5 Communication The acceptability of the risk partly depends on the

type of harm. In the case of harm to people, then the
tolerability of that risk should be lower. If the
Management of any risk requires good consequence is only commercial, such as delay or
communication, co-ordination and co-operation. lack of performance, the client and design team are
Many environmental and human disasters are entitled to consider how much they are prepared to
caused not by deliberate omission but by oversight spend to control the risk. In some cases there are
and lack of communication. Many accidents occur legal or regulatory requirements for particular types of
because of changes made by somebody who risk, but the difficulty of quantifying risk results in
didn’t understand why it was done the way it was. many laws being targeted at processes and
In many countries there is now a legal requirement behaviour rather than directly at risk.
for co-ordination of both design and construction
work. In the European Union, this stems from the Consequence is an important part of risk. Structural
Temporary or Mobile Construction Sites failure in one situation may have relatively low
Directive3.6, implemented in the UK as the consequences; in another situation a similar amount
Construction (Design and Management) of structural damage could result in much greater
Regulations3.7. This has resulted in an unfortunate harm to people, financial cost or disruption to society.
trend for co-ordination of health and safety to be As an example, consider a bridge leading to a farm;
seen as a separate responsibility from co-ordination the potential consequences of collapse into a river,
of the design itself. Structural engineers should, in onto a main road or onto a high speed railway could
order to do their job, be competent in relevant health be very different.

The Institution of Structural Engineers Risk in structural engineering 7

3.7 Principles of risk management

While probability and consequence are separate construction industry has the acronym ERIC –
aspects of risk, a subjective assessment of probability Eliminate, Reduce, Inform, Control. Further detail is
might subconsciously take the consequences into given in Section 5.3. In practice, it will often be possible
account. The relative probability of the same event to combine actions from several levels to give the
affecting different structures may also be relevant. optimum overall reduction in risk. Designers have more
Terrorist attack would not normally be a valid opportunity to eliminate hazards than constructors do.
consideration for a private housing development but Once the design is fixed, the scope for hazard
might well be for government offices. elimination is much reduced and the constructor may
only be able to reduce the risk. Other aspects of the
Account needs to be taken of people’s risk hierarchy include use of engineered measures in
perception, as well as the reality of the risk. Some preference to management systems, and controls that
risks occur naturally, while others are artificial. Gravity protect everyone rather than those that protect
loading is very predictable (the initiating event for individuals. That said, flexibility and proportionality are
failure under normal loads is usually an error of some necessary. For a one-off or infrequent activity, it may be
kind) but the public expectation is that all structures justifiable and sometimes safer to use a different
will be ‘safe’ under gravity loads so the acceptable approach. For example, when using a crane in a
risk level is very low. In contrast, there is some restricted space, it may be better to rely on skilled
tolerance to limited structural failures in UK from operators and good planning than to try to set up
severe storms and, for events such as earthquakes, slewing limits and interlocks. A second example is the
there is a general feeling that some failures are more use of trained roped access engineers on inspections
or less inevitable even though the reality is that many where this avoids building a large scaffold for a day’s
are preventable at affordable cost. work.

No society can be ‘risk free’ and there are sometimes In this Report, the term ‘mitigation’ is used to refer to
demands for more stringent regulation to control the actions which reduce the consequence of the event
risk created by others. Some types of harm are but do not prevent it. Mitigation therefore sits at the
dreaded more than others, making the risk even less bottom of the hierarchy. The use of a full body
acceptable; for example, the UK’s Health and Safety harness and energy absorbing lanyard for work at
Executive considers that the risk of work related height is an example of mitigation. By allowing a fall
cancer should be weighted more highly because to occur, however, control of the situation is lost, and
people fear this more than other causes of death3.9. unexpected consequences can occur; such as
In most countries, society is becoming less willing to breakage of a lanyard, injury or suspension trauma.
accept risk, but risk appetite in any situation is For some years, this individual ‘protection’ measure
influenced by the overall balance of cost, resources and was standard for work at height in the UK; it has now
risk from other sources. been recognised as mitigation. If possible, work
should be carried out from ground level or from
working platforms. Collective mitigation measures
such as safety nets should be considered if work at
3.7 Proportionality and ALARP height cannot be eliminated or the initial fall
prevented.

The general principle of proportionality is that most

effort should be applied where the probability of harm
and/or the magnitude of harm are greatest. 3.9 Risk aversion
The idea that risks should be reduced to ‘as low as is
reasonably practicable’ is known by its initials, Risk aversion simply means an unwillingness to take
ALARP3.10. Although a principle in UK legislation (see risk. In the context of risk management it has
Section 4.4) and a useful concept, not all legal systems developed a number of meanings.
adopt it and it does not apply where there is a more
specific legal duty, e.g. in relation to work at height. The The first is that public perception of risk may not be
wording in UK law is ‘so far as is reasonably what a strictly ‘statistical’ approach would suggest.
practicable’ (SFARP or SFAIRP) but is effectively the Some risks are less acceptable to the public than
same. others; for example, in the UK fatal accidents to rail
passengers attract more public concern than road
ALARP is a useful principle, as it means that common accidents. This means that engineers need to ensure
sense can be applied. If there is a serious hazard with clear communication of risk and to take account of
a high risk of occurrence, the engineer has the perceptions.
support of the law in making a case to do something
about it, even if it is expensive. On the other hand, if The second is that people can become unable to
the only way to reduce the risk is expensive, and the accept any risk, even where necessary to avoid a
cost is completely disproportionate to the benefit, more serious risk. This can lead to a project stalling,
ALARP is a justification for not doing it. Alternatives to or to a preoccupation with trivial risks at the expense
ALARP are typically less flexible so can result in of more serious issues being overlooked3.11. For
wasted effort. example, on a project to refurbish a railway bridge, a
‘risk assessment’ identified ‘noise from trains’ as a
hazard and proposed, in mitigation, that all personnel
should wear ear protection, which would actually have
3.8 Risk management hierarchy placed them at greater risk.

It is important to ‘step back’ occasionally, particularly

Every activity should take account of the risk if working from checklists or procedures, and look at
management hierarchy3.7. The usual form in the the ‘big picture’ by asking ‘where are the significant

8 The Institution of Structural Engineers Risk in structural engineering

 Principles of risk management 3.10

risks?’ In some cases ‘do nothing’ will carry less risk good idea. In some cases, the constructor will be
than ‘do something’. Aversion to risk should not be best placed to manage a risk; for example, adverse
allowed to stifle innovation and the freedom of weather conditions. If, however, the constructor has
designers to create the right structure for the task3.12. no effective means of controlling the risk, this will be
By studying the risks, it is often possible to find a reflected in the tendered price. The client may believe
‘win-win’ solution that reduces risk and at the same the price is fixed, but that price may be much higher
time improves the structure, making it safer, more than if the client had accepted the risk or perhaps
useful and more economical over its planned shared the burden. If the risk materialises, the history
lifetime3.13. of contract disputes suggests the expected fixed
price may not be realised. There may be
circumstances where a client is prepared to pay extra
in exchange for confidence in the price, but this
3.10 Resilience should not be done by default or without knowledge
and generally it is wiser to be completely open about
the risks.
Severe but unusual or infrequent hazards require a
cautious approach to risk. While a natural hazard While, with sufficient awareness, commercial risk may
such as a tsunami cannot be prevented, its effects be transferred to others, risks to health and safety
can be reduced by building structures that are should not be. In many cases, the law will prevent
resistant to flooding. Such events are so rare that such delegation. For example, the 2007 revision of
there is often insufficient information to make a proper the UK’s CDM Regulations3.15 places more duties on
assessment of the probability of occurrence. An the client. This recognises that the designer’s
unstructured estimate may show a low probability of opportunity to manage risk will depend on the
failure, providing a false reassurance. availability of adequate time and funds.

For low probability, high consequence events it is

better to start from the consequences, and consider
whether, under any circumstances, failure would be 3.12 Societal risks
acceptable. If not, something should be done to
reduce the risk. Examples might be the impact of a
ship on a fully loaded suspension bridge or a natural Some of the work that structural engineers do can
event such as the tsunami that struck North East expose members of the public to risk, even those
Japan in 2011 resulting in many thousands of deaths who have not chosen to be involved with the
and putting a nuclear power station into a potentially project. Failure of an industrial structure resulting in
dangerous condition3.14. an explosion or a release of toxic gas could have
serious consequences extending beyond the site or
The important aspect of resilience based design is project. The Buncefield3.16 (2005) and
that it does not involve setting a probability for the Flixborough3.17 (1974) explosions and the Bhopal3.18
load or initiating event. Where the consequences are (1984) toxic gas release did not result from structural
unacceptable, the design should eliminate the failures, but similar events could happen if supports to
possibility that the hazard can have a serious impact. process vessels failed.
This may require consideration of a ‘beyond design
basis’ event. This does not mean, however, that the Very large risks may give rise to ‘societal concern’,
design should aim to reduce the risk, or even the meaning risks that could impact on society and could
consequences, to zero. A tsunami would be have adverse repercussions on those responsible for
expected to result in significant damage to putting in place the high level provisions and
infrastructure, but gross loss of life should be avoided arrangements for protecting people, e.g. Parliament
and the infrastructure should be repairable. or the Government of the day. The type of risks that
could provoke a socio-political response would
involve widespread or large scale detriment or the
occurrence of multiple fatalities in a single event.
3.11 Ownership and control of risks Society’s concerns are not necessarily rational; risks
that are mathematically or scientifically quite small
may provoke serious public concern. Typical
An ‘owner’ (i.e. a person or organisation that is examples relate to nuclear power generation or
responsible for managing that risk) should be railway travel.
identified for every risk. Unless there is a legal
requirement, ownership is typically allocated to the
party best able to manage the risk, although who that
is may be a matter of opinion. 3.13 Human failure and accidents
The ownership of risk typically varies in different
communities around the world. The responsibility and Many accidents that are attributed to ‘human error’
accountability for risk management may lie with the are actually the result of badly designed systems that
professional engineer, or the developer and their put humans into situations where they are more likely
team, or with the ministries of the state, or with those to make mistakes.3.19 HSG483.20 gives examples of
bodies notionally independent of actual design and accidents in the transport, process and nuclear power
construction who check for compliance to laws, industries. Failures can occur at any level in an
codes and regulations. organisation, from the shop floor to senior
management. For many of these major accidents,
Some clients choose to use contracts to place human failure was not the sole cause but one of a
commercial risk with the constructor, giving the client number of causes, including technical and
the ‘benefit’ of a fixed price. This is not necessarily a organisational failures, that led to the final outcome.

The Institution of Structural Engineers Risk in structural engineering 9

3.14 Principles of risk management

Many ‘everyday’ minor accidents and near misses also and during the final flight of the space shuttle
involve human failures. James Reason3.21 has defined Columbia (2003), evidence that Columbia was at
five root causes of accidents, at least three of which risk was not taken sufficiently seriously. Engineers
relate primarily to human error. James Reason also with safety concerns were intimidated by
developed the ‘Swiss cheese’ or ‘failure window’ bureaucratic systems or managers who believed the
concept. This postulates multiple barriers against risk was small and were more concerned about
failure, but each barrier contains holes that represent cost and schedule. The official report3.24, 3.25 states
errors. If all the holes are aligned, or if everybody relies ‘‘... engineering teams were held to the usual
on somebody else to prevent the failure, the barrier quantitative standard of proof. But instead of having
fails. to prove it was safe to fly, they were asked to prove
that it was unsafe to fly.’’
Within the field of structural engineering, there are
many situations where design decisions can set up Following the explosion at the Texas City oil refinery
the potential for mistakes, or even create latent (2005)3.26, official criticism of the operator extended
errors. For example: beyond the circumstances of the actual incident to
– Steel columns, identical except for the grade of condemnation of their safety culture and the standards
steel, may be confused on site. at their other US refineries.
– Complex setting out arrangements are both difficult
to execute and to check. An organisation with a strong ‘safety culture’ will
– Concrete sections of different size and/or usually find it easier to have an open discussion
reinforcement at every grid line have the potential about risk, and reach more ‘honest’ conclusions
for error. about which risks are significant. The best way to
assess risks is to feel involved with the outcome, and
The so called ‘Murphy’s law’ states that if something not to assess them just to meet legal or regulatory
can go wrong, it will go wrong; the only way to avoid requirements.
this is to anticipate mistakes and design them out.
Simplicity is helpful; it is more difficult to identify errors
in complex or novel designs. It is surprisingly easy to
make apparently ludicrous mistakes such as building
structures the wrong way round or a metre out of 3.15 Prescription versus engineering
position. In a major UK industrial facility, a serious judgement
error was narrowly averted when it was realised that
the site datum for two adjoining facilities differed by a
metre. The objective of design codes is to ensure that
engineers are well informed about how design
In its 16th report, in 20073.22, the Standing Committee should be done. At the same time, they reduce the
on Structural Safety (SCOSS) used the generic need for engineers to understand the fundamental
headings of people, process and product – ‘the 3Ps’ – principles and discourage initiative and innovation.
to categorise ongoing issues. ‘People’ is the most Some countries make compliance with design
central of these, since people develop and use both codes a legal duty, while others treat them as
processes and products. guidance. There will always be situations which are
not covered by the codes, and in which the
structural engineer will need to use judgement. This
is much more than opinion; judgement should be
3.14 Safety culture supported by logical argument and the engineer’s
peers should be convinced by this.

Risk management in structural engineering requires For design of simple structures, compliance with
an understanding of how organisations behave, as routine design codes should result in an adequate
well as structures. Failure to understand any element structure, although the safety of those constructing,
of this complexity may lead to risks being higher than maintaining and using it will still need to be
expected. considered. As codes evolve, however, it is important
to keep sight of the bigger picture and consider
‘Safety culture’ describes the way an organisation issues which are not included in the codes. The
and its members respond to the need to ensure refurbishment or demolition of even simple structures
health and safety. It has been defined as ‘‘the product is likely to require judgement. There are many
of individual and group values, attitudes, perceptions, structures where a purely prescriptive approach is
competencies, and patterns of behaviour that insufficient and structural engineers have a
determine the commitment to, and the style and professional obligation to understand the background
proficiency of, an organisation’s health and safety to the notion of ‘safety’ and to take active steps to
management’’3.23. promote it. This applies particularly when designing
structures of any size for hazardous industries or that
To develop a safety culture requires leadership, may be subjected to rare but potentially catastrophic
communication, employee involvement and in events.
particular the establishment of a ‘learning culture’, not
a ‘blame culture’, within the organisation. If staff, It is therefore essential for engineers to take the time
rather than the system, are blamed for accidents they to study the fundamentals of their art and what
will tend to cover up any that occur. This will distort constitutes good practice, not merely follow rules
the picture and make it difficult to achieve a genuine blindly. This requires training and good
reduction in risk. understanding, partly because the solutions may not
be prescriptive but instead require the exercise of
A poor safety culture has been responsible for many considerable judgement. This is reinforced in a report
errors in risk management. For example, prior to by the UK’s Engineering Council3.12.

10 The Institution of Structural Engineers Risk in structural engineering

 Principles of risk management 3.16

3.16 The safety case 3.18 References

The concept of a ‘safety case’ is often applied to high 3.1 Haefeli, K. et al. Perceptions of the cost implications of
risk industries. It is only likely to be directly relevant to health and safety failures, Research Report 403.
structural engineers working in those industries, but it Sudbury: HSE Books, 2005. Available at: http://www.
provides another way of looking at risk management hse.gov.uk/research/rrpdf/rr403.pdf [Accessed:
and may stimulate thought. 18 February 2013]

The traditional approach to safety regulations in the 3.2 Health & Safety Executive. The cost to Britain of
UK was prescriptive; rules were laid down about workplace accidents and work related ill health in
what was and was not acceptable practice. In the 1995/96. HSG101. 2nd ed. Available at: http://www.
1970s there was a move towards a goal-setting hse.gov.uk/pubns/priced/hsg101.pdf [Accessed: 26
approach; employers became responsible for Oct 2012]
deciding how to manage their operations to meet
specified standards. This aligned with the increasing 3.3 National Commission on Terrorist Attacks on the United
complexity in industries such as nuclear power, and States. The 9/11 Commission Report. Available at:
the concept of a safety case was developed. http://govinfo.library.unt.edu/911/report/911Report.pdf
[Accessed: 26 February 2013]
The safety case is a document or set of documents
recording a formal assessment that may include 3.4 Taleb, N.N. The Black Swan: the impact of the highly
analysis of hazards, risks, protection and mitigation. It improbable. Rev ed. London: Penguin, 2010
is produced by the operator to assist with safety
management of a facility or system and may be 3.5 EEFIT. The Hyogo-Ken Nanbu (Kobe) earthquake of
required by, and thus submitted for approval to, a 17 January 1995: a field report. Available at:
regulatory body such as the Health and Safety http://www.istructe.org/downloads/resources-centre/
Executive. It allows the logic (‘why this is safe’) to be technical-topic-area/eefit/eefit-reports/hyogo-ken-
set down for review; it should always be the objective nanbu-kobe.aspx [Accessed; 18 February 2013]
of the operator to ensure that a good safety case is
produced, without relying on the regulator. 3.6 Council Directive 92/57/EEC of 24 June 1992 on the
implementation of minimum safety and health
The concept has expanded to include railways and requirements at temporary or mobile construction sites
offshore oil and gas. While the detailed content of a (eighth individual Directive within the meaning of
safety case is a matter for its authors, the format and Article 16 (1) of Directive 89/391/EEC). Available at:
controls on the production and use of a safety case http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=
may be subject to regulatory guidance3.27. CELEX:31992L0057: EN:HTML [Accessed: 18 February
2013]

3.7 CITB – ConstructionSkills. ‘Section 2: Hazard elimination

3.17 Conclusions and recommendations and risk reduction ‘. In CITB – ConstructionSkills. The
Construction (Design and Management) Regulations
2007 – Industry guidance for designers. King’s Lynn:
Risk management is a fundamental part of ConstructionSkills, 2007. Available at: http://uk.sitestat.
engineering and it may be argued that it should not com/citb/cskills/s?search.CDM_Designers4web_07_
require separate consideration. It is, however, an tcm17-4643&ns_type=pdf&ns_url=http://www.cskills.
essential engineering skill and needs specific org/uploads/CDM_Designers4web_07_tcm17-4643.pdf
consideration as projects become more complex. [Accessed: 18 February 2013]
Structural engineering failures can have very serious
consequences3.28. As codes of practice become more 3.8 Hall, Stephen S. ‘Scientists on trial: at fault?’, Nature,
prescriptive, the engineer’s vision can become 477, 264-269 (2011) [Online]. Available at: http://
narrower. Engineers should understand: www.nature.com/news/2011/110913/full/477264a.
– When to use engineering judgement. html [Accessed: 22 July 2013]
– What is an acceptable level of risk.
– How probable is a hazard to be realised. 3.9 Health & Safety Executive. Reducing risks, protecting
– Risk tools such as hierarchy, qualitative probability people: HSE’s decision-making process. Sudbury: HSE
and proportionality. Books, 2001. Available at: http://www.hse.gov.uk/risk/
– Risk aversion and ownership of risk. theory/r2p2.pdf [Accessed: 18 February 2013]
– Organisational factors such as safety culture.
3.10 Health and Safety Executive. ALARP ‘at a glance’.
Structural engineers (both designers and Available at: http://ww.hse.gov.uk/risk/theory/
constructors) are in a unique position to understand alarpglance.htm [Accessed: 18 February 2013]
the most significant risks in a project. This is not
limited to risks arising from the structure itself and the 3.11 Hackett, J. ‘Unintended consequences’. Judith
structural engineer’s wider view of the project can Hackitt’s blog. Available at: http://www.hse.gov.uk/
help manage risks arising from, for example, news/judith-risk-assessment/consequences211112.
mechanical services. The structural engineer should htm [Accessed: 18 February 2013]
be prepared to take a lead role in explaining the risks
to others and insist that all risks receive proper 3.12 Engineering Council. Guidance on risk for the
consideration. In particular, a flawed concept design engineering profession. Available at: http://www.engc.
should not be accepted. Younger engineers may find org.uk/risk [Accessed: 18 February 2013]
this intimidating when dealing with project managers
and clients and if necessary should ask for and 3.13 Powderham, A.J. ‘Safety as a driver for innovation in
receive support from their senior managers. design and construction of underground structures’,

The Institution of Structural Engineers Risk in structural engineering 11

3.19 Principles of risk management

Proc. International Conference on Deep Excavations, 3.28 Gilbertson, A. et al. Guidance on catastrophic events in
Singapore, 2008 construction. C699. London: CIRIA 2011

3.14 Office for Nuclear Regulation. Japanese earthquake

and tsunami: implications for the UK nuclear industry.
Final report. ONR-FR-REP-11-002. Available at: http:// 3.19 Bibliography
www.hse.gov.uk/nuclear/fukushima/final-report.pdf
[Accessed: 18 February 2013]
Blockley, D. et al. ‘Infrastructure resilience for high-impact
3.15 The Construction (Design and Management) low-chance risks’. ICE Proceedings, Civil Engineering Special
Regulations 2007 (SI 2007/320). Available at: http:// Issue, 165(CE6), November 2012, pp13-19
www.opsi.gov.uk/si/si2007/20070320.htm [Accessed:
18 February 2013] Hudson, S. et al. ‘Engineering resilient infrastructure’. ICE
Proceedings, Civil Engineering Special Issue, 165(CE6),
3.16 Explosion Mechanism Advisory Group report. London: November 2012, pp5-12
Buncefield Major Incident Investigation Board, 2007.
Available at: http://www.buncefieldinvestigation.gov.uk/ Lord Cullen. The Ladbroke Grove Rail Inquiry. Part 2 report.
reports/buncefieldagr.pdf [Accessed: 18 February 2013] Sudbury: HSE Books, 2001. Available at: http://www.
railwaysarchive.co.uk/documents/HSE_Lad_Cullen002.pdf
3.17 The Flixborough disaster: report of the Court of Inquiry. [Accessed: 13 September 2010]
London: HMSO, 1975
Montgomery, M. et al. ‘An innovative approach for improving
3.18 Srishti. Surviving Bhopal 2002. Toxic present – toxic infrastructure resilience’. ICE Proceedings, Civil Engineering
future: a report on human and environmental chemical Special Issue, 165(CE6), November 2012, pp27-32
contamination around the Bhopal disaster site. New
Delhi: Srishti, 2002. Available at: http://www.bhopal. Neale, B. ‘Introduction to infrastructural resilience’. ICE
net/oldsite/documentlibrary/survivingbhopal2002.doc Proceedings, Civil Engineering Special Issue, 165(6), November
[Accessed: 18 February 2013] 2012, pp3-4

3.19 Williams, R. This one will bring the house down: ICE
IStructE HSE SCOSS Prestige Lecture 28th April 2009.
Available at: http://cms.structural-safety.org/assets/
uploaded//documents/118_SC09.039%20-
%20Prestige%20Lecture%202009%20-
%20Richard%20Williams.pdf [Accessed: 18 February
2013]

3.20 Health & Safety Executive. Reducing error and

influencing behaviour. HSG48. 2nd ed. Sudbury: HSE
Books, 1999. Available at: http://www.hse.gov.uk/
pubns/priced/hsg48.pdf [Accessed: 18 February 2013]

3.21 Reason, J.T. Managing the risks of organizational

accidents. Aldershot: Ashgate, 1997

3.22 Standing Committee on Structural Safety. 16th Biennial

report. Available at: http://www.structural-safety.org/
biennialreport [Accessed: 18 February 2013]

3.23 Advisory Committee on the Safety of Nuclear

Installations. ACSNI Study Group on Human Factors
third report: Organising for safety. Sudbury: HSE
Books, 1993

3.24 Columbia Accident Investigation Board. The Report.

Available at: http://www.nasa.gov/columbia/caib/html/
VOL1.html [Accessed: 18 February 2013]

3.25 Mason, R.O. ‘Lessons in organisational ethics from the

Columbia Disaster: can a culture be lethal?’
Organizational Dynamics, 33(2), 2004, pp128-142

3.26 Fatal Accident Investigation Report: Isomerization Unit

Explosion Final Report. Available at: http://www.bp.
com/liveassets/bp_internet/us/bp_us_english/STAGING/
local_assets/downloads/t/final_report.pdf [Accessed:
18 February 2013]

3.27 Guidance on the purpose, scope and content of

nuclear safety cases. T/AST/051, Issue 001. Available
at: http://www.hse.gov.uk/nuclear/operational/tech_
asst_guides/tast051.pdf [Accessed: 18 February
2013]

12 The Institution of Structural Engineers Risk in structural engineering

4 Legal background

4.1 Introduction Regulations contain more detail, so need to be easier

to change. The act, and the general duties under, for
example, The Management of Health and Safety at
This chapter discusses UK law4.1 except where stated Work Regulations 19994.4 are goal setting and leave
otherwise. Many of the broad principles, however, employers freedom to decide how to control risks that
apply in other countries, although the law itself may be they identify. Regulations identify some risks specifically
written very differently. A summary of the legal and set out specific action that must be taken.
background in Europe, the United States and Hong Sometimes, these requirements are absolute, i.e. there
Kong is given to illustrate this. Whichever country is is a need to do something without qualification.
being worked in the engineer must be aware of the
local legislative framework, particularly in terms of An approved code of practice (ACoP) offers practical
personal or corporate liability. examples of good practice. It gives advice on how to
comply with the law by, for example, providing a
guide to what is ‘reasonably practicable’. For
example, if regulations use words like ‘suitable’ and
4.2 Law as it affects structural ‘sufficient’, an ACoP can illustrate what this requires
in particular circumstances. An ACoP has a special
engineering legal status. If employers are prosecuted for a
breach of health and safety law, and it is proved that
they have not followed the relevant provisions of the
Most legislation which affects structures is about
ACoP, a court can find them at fault unless they can
preventing people from being harmed or being
show that they have complied with the law in some
made ill; the law says people must not put
other way.
themselves, other workers or the public in
danger4.2. Both construction work and the stability of
Guidance is available on a range of subjects. Some
completed structures are subject to legal
is specific to the health and safety problems of an
requirements. Damage to the environment is also
industry or of a particular process used in a number
becoming increasingly regulated. Health and safety
of industries. The main purpose of guidance is to
law (criminal legislation) applies to all businesses,
interpret what the law says, to help people comply
however small. It covers employees (full or part-time,
with the law and to give technical advice. Following
temporary or permanent) and the self-employed.
guidance is not compulsory and employers are free
Controlling danger at work is no different from any
to take other action. But if they do follow guidance
other task; in other words, health and safety needs
they will normally be doing enough to comply with
specific action to manage it. The structural engineer
the law.
needs to recognise problems, know enough about
them, decide what to do and act on the solutions. It is
not just highly unusual or exceptional circumstances
that cause accidents or ill health. Some basic thought
and action beforehand can usually prevent them. Civil 4.4 Reasonably practicable
law is concerned with allocating responsibility for loss
typically in relation to a contract or alleged negligence.
While criminal law is based on statutes, i.e. laws What is meant by ‘reasonably practicable’? It may be
made by parliament, civil law is mainly based on expressed as, ‘so far as is reasonably practicable’
precedent. Health and safety does not have to be (SFAIRP) or, ‘as low as reasonably practicable’
expensive, time consuming or complicated, but (ALARP). SFAIRP is the term most often used in the
engineers may be culpable if they do not keep Health and Safety at Work etc. Act4.3 and in
themselves informed over what causes harm. regulations. ALARP4.5 is the term used by risk
specialists. In the view of the Health and Safety
Executive (HSE), the two terms are interchangeable
(except when drafting formal legal documents, when
4.3 Acts, regulations, guidance and the correct legal phrase must be used).

ACoPs The definition set out by the Court of Appeal4.6 is:

‘‘‘Reasonably practicable’ is a narrower term than
‘physically possible’... a computation must be
The basis of British health and safety law is the
made by the owner in which the quantum of risk is
Health and Safety at Work etc. Act 19744.3 (HSWA,
placed on one scale and the sacrifice involved in
the Act4.3). An act is primary law, enacted by
the measures necessary for averting the risk
parliament. The Act sets out the general duties that
employers have towards employees and members of (whether in money, time or trouble) is placed in the
the public, and that employees have to themselves and other, and that, if it be shown that there is a gross
to each other. These duties are qualified in the Act by disproportion between them – the risk being
the principle of ‘so far as is reasonably practicable’ (see insignificant in relation to the sacrifice – the
Section 4.4). defendants discharge the onus on them.’’

Regulations are secondary law, approved by a In essence, making sure that a risk has been
minister under powers made under the act. reduced to ‘as low as reasonably practicable’ is

The Institution of Structural Engineers Risk in structural engineering 13

4.5 Legal background

4.5 Practicable

Risk must be reduced unless there Unacceptable The words ‘reasonably practicable’ should not be
are exceptional reasons why not confused with ‘practicable’. In a legal context,
‘practicable’ infers a statutory obligation that has to
be met if, in the light of current knowledge, it is

Increasing risk
feasible (irrespective of cost or difficulty). Put at its
HSE may require evidence simplest, ‘practicable’ means ‘if it can be done, it
that risk is ALARP Tolerable must be done’.

4.6 Burden of proof

ALARP applies but evidence Broadly
is not usually required acceptable Where a duty-holder is required to do what is
‘reasonably practicable’ or ‘practicable’ to achieve a
safe system of work, Section 40 of the Act4.3
provides that the burden is on the defendant to satisfy
the court that it was not practicable or reasonably
Figure 4.1 Framework for tolerability of risk practicable to do more to control the risk than was in
fact done. This is often referred to as a ‘reverse
burden’, because it reverses the normal situation that
the prosecution must prove the facts beyond
reasonable doubt.
about weighing the risk against the sacrifice needed
to further reduce it. The decision is loaded in favour
of health and safety as the presumption is that the
duty-holder, i.e. the employer, should implement the 4.7 Liability under civil law and duty of
risk reduction measure. To avoid having to make care
this ‘sacrifice’, the duty-holder must be able to
show that it would be grossly disproportionate to
the benefits of risk reduction that would be As well as criminal law, those who are responsible for
achieved. Thus, the principle behind the process is harm to others may be sued for damages under civil
not one of balancing the costs and benefits of law. Liability may arise from the terms of a contract,
measures but, rather, of adopting measures except or may exist irrespective of contract under the ‘duty
where they are ruled out because they involve of care’ principle.
grossly disproportionate sacrifices. Figure 4.14.7
illustrates how this principle is applied across a range Duty of care is the obligation to exercise a level of
of risk severity. care towards an individual, as is reasonable in all the
circumstances, to avoid injury or loss of property to
ALARP should be applied to decide whether a that individual. It is therefore based upon the
proposed risk management measure is necessary. relationship of the parties, the negligent act or
This may require an assessment of the initial risk, omission and the ability to reasonably foresee loss to
and then the residual risk after applying the selected that individual. A negligent act is an unintentional but
measures (e.g. elimination, substitution, mitigation, unreasonably careless act that results in loss. Only a
etc.) and comparing the reduction in risk with the negligent act will be regarded as having breached a
cost of providing the measures. Such cost benefit duty of care. Duty of care arises from the precedent
assessments can be time consuming and can be set by previous judgements, rather than law enacted
impractical where, as is often the case in structural by parliament. Liability for breach of a duty of care
engineering, data is not available. As an alternative, can therefore depend on the public policy at the time
it can be assumed that following accepted and the case is heard.
relevant ‘good practice’ will ensure that risks are as
low as reasonably practicable. In effect, this relies The law of negligence condemns as negligent any act
on assessments previously carried out by others, or omission that falls short of a standard to be
which have resulted in the body of knowledge expected of ‘the reasonable person’. The application
called ‘good practice’. It is important to ensure that of this test by the courts depends on the type of
the precedent is relevant; what is good practice for case. In a clinical negligence action the standard was
a farm building may not be good practice in a multi- defined in the ‘Bolam test’4.10 (1957). This set out the
storey residential block, and good practice for the test used when a judge is considering whether or not a
residential block may not apply to a hazardous doctor has been negligent, and has subsequently been
industrial facility. extended to other professions.

These requirements can appear very complex and The case held that a doctor is not in breach of the
daunting but, in reality, industry norms, good practice duty of care, ‘‘if he has acted in accordance with a
and a professional approach will guide designers practice accepted as proper by a responsible body of
through the process. However, underlying this is an medical men skilled in that particular art’’. The
assumption that judgements4.8 are made by practical effect of the test is that a judge will hear
competent persons. For high hazards, complex or evidence from experts in the appropriate speciality
novel situations, good practice can be built upon using and must decide whether the actions of the doctor
more formal decision making techniques, including were proper. Often, there are several acceptable
cost-benefit analysis4.9. ways of doing something and compliance with any of

14 The Institution of Structural Engineers Risk in structural engineering

 Legal background 4.8

these will mean that there was no breach of duty of building design and construction must comply in
care. Naturally, experts often disagree over these the interests of the health and safety of building
issues and the judge must decide whose evidence is users, of energy conservation, and of access to
to be preferred. and use of buildings. The requirements are often
referred to as ‘functional’ and are expressed in
It is important that anyone reviewing a case as an terms of what is ‘reasonable’, ‘adequate’ or
expert, or giving an informal view, understands the ‘appropriate’.
Bolam test. The fact that the person giving an opinion
would not have done things in the same way does Practical guidance on ways to comply with the
not automatically mean that there was a breach of functional requirements is contained in Approved
duty of care. The actions taken may be acceptable to Documents, or a Technical Handbook. These
‘a responsible body of opinion’ and research (such as contain general guidance on the performance
a literature search) may be needed to check the expected of materials and building work. There is a
position. When considering whether one owes a duty legal presumption that if the guidance has been
of care, up to date information is essential, as case followed, then this is evidence that the work has
law evolves over time. complied with the Building Regulations. It is,
however, quite acceptable to use alternative
methods of compliance provided they fully satisfy
the regulations. A designer may put forward other
4.8 Law enforcement ways of meeting the regulations, but these will
require approval. This will be particularly important
when assessing the level of risk in modifications or
In the UK, the HSE and local government are generally changes of use to existing buildings that do not
the enforcing authorities for health and safety meet the current regulations.
regulation4.11. The HSE’s mission is to protect people’s
health and safety by ensuring risks in a changing In addition, Section 4A of the Workplace (Health,
workplace are properly controlled. It looks after health Safety and Welfare) Regulations4.16 requires that a
and safety in factories, farms, mines, nuclear ‘‘building shall have a stability and solidity
installations, hospitals and schools, offshore gas and oil appropriate to the nature of the use of the
installations, the gas grid, the movement of dangerous workplace’’ and applies to any workplace
goods and substances and many other aspects of the irrespective of the regulations in force during its
protection both of workers and the public. Local original construction.
authorities are responsible for enforcement in offices,
shops and other parts of the services sector.

The consequence of failure has increased in profile 4.11 Europe

over recent years. Since the Corporate Manslaughter
and Corporate Homicide Act 20074.12, companies
and organisations can be found guilty of corporate Across Europe the work of the European
manslaughter as a result of serious management Commission is improving the coherence and
failures resulting in a gross breach of a duty of care. In rigour of issues concerning the safety of people in
addition, the Health and Safety (Offences) Act 20084.13 their environment. There are, however, deeper and
has increased penalties and provides courts with persistent differences in collective and individual
greater sentencing powers for those who flout health responsibility, chain of responsibility, duty to act,
and safety legislation. the necessity of insurance and its workings, and
the balance between prescription and
goal-setting.

4.9 Designer’s role Each member state is required to pass legislation

that implements each directive in that state,
although the details of implementation may vary. In
In the UK, the designer’s duties are defined most European countries the law requires
specifically in the Construction (Design and employers to protect workers against all risks. As it
Management) Regulations 20074.14. A designer is is not possible to be absolutely safe (otherwise, for
anyone who carries out design work as part of a example, nobody could drive on the roads for a
business, including temporary works, fabrication living) the courts apply these laws with a view to
details and details by the constructor. This can include what is reasonable. Courts in the UK, however,
the client. The term ‘design’ is a wide term, and interpret the law precisely according to precedent
includes drawings, calculations, design details, and statutes. Accordingly, the concept of ‘so far as
specifications and bills of quantity. The designer must is reasonably practicable’ has developed.
not only design the work to be safe, so far as is
reasonably practicable, but also has an absolute duty European directives have prompted much of the
to be competent and to ensure that the client is aware current UK legislation on health and safety. The CDM
of their duties. Similar laws apply throughout the Regulations4.14, for example, were enacted as a result
European Union (see Section 4.11). of the Temporary or Mobile Construction Sites Directive
(92/57/EEC)4.17, which sets out minimum safety and
health requirements for construction sites within the
European Union.
4.10 Building Regulations
In France, the responsibility for avoiding disasters (not
necessarily risk, or accidents) is varied. It lies with the
In the UK, Building Regulations4.15 set out state to initiate understanding of issues like natural
‘requirements’ with which the individual aspects of and unnatural events, landslides, flash floods and the

The Institution of Structural Engineers Risk in structural engineering 15

4.12 Legal background

acceptability of industrial processes near the and to complete a kind of risk assessment. The code
population. It is with the population as a whole for contains a table of critical stages of construction so that
issues like the protection of the water cycle, for which enhanced supervision is imposed. Examples include
there is a specific law. It is with the constructor of demolition of complex structures, such as flat slabs,
buildings and it is with any company director, by pre-stressed concrete, transfer plates, hangers, long
default. The role of the professional engineer span beams greater than 10m, steel-framed
disappeared at the French Revolution. Strangely construction and cantilevered structures over streets
enough, in a country with such elegantly drafted and with spans greater than 1.2m.
complete legal codes, it is necessary to study the
effect of the law in practice to understand what
society expects from its constructors.
4.13 United States
In a case in France, impending at the time of writing,
the courts will be discussing how a disaster occurred
when there were three independent structural In the United States (US), regulation of building
engineering studies during design and construction. design is delegated to each state. Most states then
These were by the Maitrise d’Oeuvre (broadly, the delegate regulation to local government, i.e. cities,
project manager), the constructor and finally an counties and townships. Cities and counties with
independent checker for code compliance. The large populations generally police design well.
details of the case are still sub judice, but it appears However, jurisdictions with smaller populations may
that there were several missed opportunities to have no requirements.
prevent the disaster. The precursors to failure should
have been visible in both the procedural and the In addition to local ordinances to enforce building
structural systems. codes, each state and the six territories licence
design professionals through the police powers
granted by the US Constitution. The licensing
process requires that design professionals use an
4.12 Hong Kong appropriate ‘standard of care’ when preparing
construction documents. Courts generally interpret
this to mean that design professionals should follow
As in other former British colonies such as Australia, model codes and standards for construction.
much of Hong Kong’s safety and health legislation is
based on UK practice. Safety and health law is For construction site safety, the federal rules put
basically criminal law, administered through the together by the Occupational Health and Safety
magistrates’ courts. The magistrate has the power, Administration (OSHA) apply. These ‘minimum’
on conviction, to impose fines on companies or requirements are intended to provide a safe work
employees and can, in extreme cases, impose place. In some cases, the OSHA rules supersede
custodial sentences. Civil law comes into safety building codes. However, in most cases they cover
issues through contract and tort, and the latter is the areas of means and methods not addressed by codes.
basis on which personal injury claims are brought.
Construction disputes are settled in civil courts. In
Hong Kong’s Occupational Safety and Health some cases, construction accidents are addressed in
Ordinance 19984.18 is similar to the UK’s HSWA4.3, and a criminal court. However, while criminal courts in the
includes the same concept of ‘reasonably practicable’. US seldom punish design professionals,
compensation awards in civil courts can be ruinous.
The Construction Sites (Safety) Regulations 19784.19
have many similar provisions to those contained in UK
legislation. Both sets of regulations impose a duty on
employers to ensure the safety of their employees, 4.14 Summary and conclusions
whether or not a potential hazard is created by the
employer or others, and they also have a duty to any
other person who may be affected by the construction – The legislation affecting structural engineering risk
works. Employees also have obligations to comply with controls:
the regulations, the most obvious being to wear – Health and safety at work (preventing people
personal protective equipment as specified. from being harmed or being made ill).
– Stability and safe maintenance of structures, also
Private sector buildings are subject to the control of to protect people.
the Buildings Department. The Technical – Protection of the environment.
Memorandum for Supervision Plans 20094.20 is legally
enforceable and lays down specific requirements for – The law may not significantly influence the decisions
site supervision. This is supported by the guidance in an engineer makes, but generally requires a record
the Code of Practice for Site Supervision 20094.21. The of the reasons for those decisions.
required number of Technical Competent Persons, the
frequency of their inspections and their qualifications – Controlling danger at work is no different from any
and experience must be calculated using a formula other task. The structural engineer needs to
based on the scale, complexity and risk of the project. recognise problems, know enough about them,
Typically, the most senior will be a qualified engineer decide what to do and act on the solutions.
who is recognised as a registered professional
engineer (structural, geotechnical or civil as – Whichever country is being worked in, the engineer
appropriate). In the most complex geotechnical must be aware of the local legislative framework.
projects, director grade staff from the consultant must While the principles of risk management remain
visit the site on a weekly basis. The site staff are similar, the legal responsibility for applying them can
required to pay attention to particular high risk items vary substantially.

16 The Institution of Structural Engineers Risk in structural engineering

 Legal background 4.15

4.15 References see http://www.legislation.gov.uk/all?title=

building%20regulations]

4.1 Health & Safety Executive. Health and safety 4.16 The Workplace (Health, Safety and Welfare)
regulation: a short guide, HSC13(rev1). Sudbury: HSE Regulations 1992 (SI 1992/3004). Available at: http://
Books, 2003. Available at: http://ww.hse.gov.uk/pubns/ www.legislation.gov.uk/uksi/1992/3004/contents/made
hsc13.pdf [Accessed: 18 February 2013] [Accessed: 18 February 2013] as amended by The
Health and Safety (Miscellaneous Amendments)
4.2 Health & Safety Executive website. Available at: http:// Regulations 2002 (SI 2002/2174). Available at: http://
www.hse.gov.uk [Accessed: 18 February 2013] www.legislation.gov.uk/uksi/2002/2174/contents/made
[Accessed: 18 February 2013]
4.3 The Health and Safety at Work etc. Act 1974 [as
amended]. Available at: http://www.legislation.gov.uk/ 4.17 Council Directive 92/57/EEC of 24 June 1992 on the
ukpga/1974/37 [Accessed: 18 February 2013] implementation of minimum safety and health
requirements at temporary or mobile construction sites
4.4 Management of Health and Safety at Work Regulations (eighth individual Directive within the meaning of
1999 (SI 199/3242). Available at: http://www.opsi.gov. Article 16 (1) of Directive 89/391/EEC). Available at:
uk/si/si1999/19993242.htm [Accessed: 18 February http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=
2013] CELEX:31992L0057: EN:HTML [Accessed: 18 February
2013]
4.5 Health and Safety Executive. ALARP ‘at a glance’.
Available at: http://www.hse.gov.uk/risk/theory/ 4.18 Occupational Safety and Health Ordinance. Cap 509,
alarpglance.htm [Accessed: 18 February 2013] Available at: http://www.legislation.gov.hk/blis_pdf.nsf/
6799165D2FEE3FA94825755E0033E532/
4.6 Edwards v. National Coal Board, [1949] 1 All ER 743 9198BE222266C421482575EF0012128E/$FILE/CAP_
509_e_b5.pdf [Accessed: 18 February 2013]
4.7 Health & Safety Executive. Reducing risks, protecting
people: HSE’s decision-making process. Sudbury: HSE 4.19 Construction Sites (Safety) Regulations. Cap 59I.
Books, 2001. Available at: http://www.hse.gov.uk/risk/ Available at: http://www.legislation.gov.hk/blis_pdf.nsf/
theory/r2p2.pdf [Accessed: 18 February 2013] 6799165D2FEE3FA94825755E0033E532/
CB7ACD5F5F2AF7D1482575EE00356ACA/$FILE/
4.8 Institution of Civil Engineers. A review of, and CAP_59I_e_b5.pdf [Accessed: 18 February 2013]
commentary on, the legal requirement to exercise a
duty ‘so far as is reasonably practicable’ with specific 4.20 Government of Hong Kong Special Administrative
regard to designers in the construction industry. Region. Buildings Department. Technical memorandum
London: ICE, 2010. Available at: http://www.ice.org.uk/ for supervision plans 2009. Available at: http://www.
Information-resources/Document-Library/So-Far-As-Is- bd.gov.hk/english/documents/code/TMSS2009_e.pdf
Reasonably-Practicable [Accessed: 18 February 2013] [Accessed: 18 February 2013]

4.9 Health & Safety Executive. HSE principles for Cost 4.21 Government of Hong Kong Special Administrative
Benefit Analysis (CBA) in support of ALARP decisions. Region. Buildings Department. Code of Practice for
Available at: http://ww.hse.gov.uk/risk/theory/alarpcba. Site Supervision 2009. Available at: http://www.bd.gov.
htm [Accessed: 18 February 2013] hk/english/documents/code/SS2009_e.pdf [Accessed:
18 February 2013]
4.10 Bolam v Friern Hospital Management Committee, High
Court, [1957] 1 WLR 583

4.11 Health & Safety Executive website. Available at: http:// 4.16 Bibliography
www.hse.gov.uk/aboutus/index.htm [Accessed:
18 February 2013]
Confidential Reporting on Structural Safety [CROSS] website.
4.12 Corporate Manslaughter and Corporate Homicide Act Available at: http://www.structural-safety.org [Accessed:
2007. Available at: http://www.legislation.gov.uk/ukpga/ 18 February 2013]
2007/19/contents [Accessed: 18 February 2013]
Gilbertson, A. CDM 2007 – Workplace ‘in-use’’ guidance for
4.13 Health and Safety (Offences) Act 2008. Available at: designers. C663. London: CIRIA, 2007
http://www.legislation.gov.uk/ukpga/2008/20/contents
[Accessed: 18 February 2013] Gilbertson, A. CDM 2007 – Construction work sector guidance
for designers. C662. London: CIRIA, 2007
4.14 Construction (Design and Management) Regulations
2007 (SI 2007/320). Available at: http://www. Health & Safety Executive. Economic Analysis Unit (EAU)
legislation.gov.uk/uksi/2007/320/contents/made appraisal values. Available at: http://www.hse.gov.uk/economics/
[Accessed: 18 February 2013] and The Construction eauappraisal.htm [Accessed: 18 February 2013]
(Design and Management) Regulations (Northern
Ireland) 2007. Available at: http://www.legislation.gov. House of Commons. Science and Technology Committee.
uk/nisr/2007/291/contents/made [Accessed: Scientific advice, risk and evidence based policy making.
18 February 2013] Volume 1: Report, together with formal minutes. HC 900-I.
London: The Stationery Office, 2006. Available at: http://www.
4.15 The Building Regulations 2010 (SI 2010/2214). publications.parliament.uk/pa/cm200506/cmselect/cmsctech/
Available at: http://www.legislation.gov.uk/uksi/2010/ 900/900-i.pdf [Accessed: 18 February 2013]
2214/pdfs/uksi_20102214_en.pdf [Accessed: 7
March 2013] [Note that similar but separate House of Lords. Select Committee on Economic Affairs.
regulations apply in Scotland and in Northern Ireland, Government response to the management of risk. HL Paper

The Institution of Structural Engineers Risk in structural engineering 17

4.16 Legal background

249. London: The Stationery Office, 2006. http://www.

publications.parliament.uk/pa/ld200506/ldselect/ldeconaf/249/
249.pdf [Accessed: 18 February 2013]

House of Lords. Select Committee on Economic Affairs.

Government policy on the management of risk. Volume I:
Report. HL Paper 183-1. London: The Stationery Office, 2006.
Available at: http://ww.publications.parliament.uk/pa/ld200506/
ldselect/ldeconaf/183/183i.pdf [Accessed: 18 February 2013]

Iddon, J. and Carpenter, J. Safe access for maintenance and

repair: guidance for designers. C686. 2nd ed. London: CIRIA,
2009

Institute of Doctors and Health & Safety Executive. Leading

health and safety at work: leadership actions for directors and
board members. INDG417. Available at: http://www.hse.gov.uk/
pubns/indg417.pdf [Accessed: 18 February 2013]

Rail Safety and Standards Board. Safety Decisions Programme:

the route to ‘Taking Safe Decisions’. London: RSSB, 2007.
Available at: http://www.rssb.co.uk/SiteCollectionDocuments/pdf/
vtsic_presentations/RouteToTakingSafeDecisions.pdf [Accessed:
18 February 2013]

Wright, I. ‘Risk and liability for the structural engineer: a legal

perspective’. The Structural Engineer, 81(14), 15 July 2003,
pp23-35

18 The Institution of Structural Engineers Risk in structural engineering

5 How to manage risk

5.1 Introduction Eliminate

First consideration should be given to avoiding the
hazard in the first place:
This chapter describes methods and approaches for – by elimination (remove the hazard that is the cause
managing risk. Like Chapter 3, it is written from a of the risk); or
structural engineering perspective, is not intended to – by substitution (replace the hazard with something
be specific to any country and is not a definition of less dangerous).
legal requirements. Except where noted otherwise,
this guidance can be applied by structural engineers As at all stages in risk management, action should be
who are responsible for any stage of the structure’s proportionate to the risk. There is little value in
life cycle. eliminating a relatively minor or easily managed
hazard at disproportionate cost, or if doing so creates
further hazards.

5.2 First identify the hazards Reduce

If the hazard cannot reasonably be avoided the
risk should be reduced, by those responsible,
The first step in risk management is to identify the using one or more of the following, in order of
hazards. Particular effort should be put into identifying preference:
the project specific and unusual hazards. Some of (1) Exposing people to less of the hazard (e.g.
the most serious incidents have occurred because a reduce the amount of work at height).
hazard was not identified. Depending on the type of (2) Measures that protect everyone (e.g. provide
structure and the surrounding environment, for working platforms).
example, if the structure will contain hazardous (3) Physical barriers between people and hazards.
materials or be at risk from impact of vehicles, other (4) Mitigation (reduce the consequences if the risk
people and professions should be involved. More occurs, e.g. provide safety nets).
formal approaches to hazard identification may be (5) Personal protective equipment (PPE).
appropriate for complex structures or environments,
and are discussed later in this chapter. The flowchart In any area with limited access, measures should be
in Figure 12 of Reference 5.1 may be found useful. considered to facilitate evacuation and rescue in the
event of an incident or incapacity, and these should
To identify less obvious hazards, an engineer might take account of any remaining hazards.
ask ‘what is unusual about this project?’ or ‘what
could go wrong?’ In this context, a hazard is not just Inform and Control
something that might harm people, but also If it is not feasible to eliminate or reduce risk,
something that could seriously disrupt the project. information about the risk should be passed on so
that the risk can be controlled. Where responsibility
Hazards may be intrinsic to the project, such as for the process passes between parties, such as
building over water, or may be introduced by the from designer to constructor or constructor to user,
design, the construction method or the use of the the first party should inform the second about any
structure. Hazard identification is not, therefore, a risk which would be unusual, difficult to manage or
one-off task, but should be repeated by the relevant would not be obvious. Typically, a designer will
engineer as the structure progresses through its life apply the first three stages of ERIC but will have no
cycle. control over site activity. Subject to the constraints
of the design, which may limit the opportunity for
The use of pre-formulated checklists can provide a elimination, a constructor can apply all four stages.
false sense of completeness; brainstorming the A user of the structure may only be able to apply
hazards from scratch is less likely to miss an unusual controls.
hazard or a particular vulnerability.
Finally, if the risk cannot be reduced to a low level,
consider whether the proposed activity is worth the
risk. This may mean going back to the conceptual
5.3 Apply the risk management hierarchy stage and approaching the whole project, design or
construction sequence in a different way.

There is a broadly accepted hierarchy for managing It is important to emphasise that good risk managers
hazard and risk to people, with removal of the hazard will often apply these measures subconsciously and
as the first priority and mitigating the consequences automatically; for example, by never introducing a
as the last. There are several variants, used in hazard in the first place. The objective of risk
different industries and in different countries. management is not to score points by demonstrating
how much risk has been removed; rather, it is to
Eliminate, reduce, inform, control (ERIC) has been scrutinise the project to see whether any more risk
recognised as best practice for construction work in can reasonably be removed.
the UK5.1, 5.2. For structural engineering, the following
hierarchy is recommended, subject to any local The opportunity to apply different types of risk
regulations: reduction is shown in Figure 5.1. This is only

The Institution of Structural Engineers Risk in structural engineering 19

5.4 How to manage risk

Opportunity
Key
To eliminate
To reduce
To control

Concept Detailed Construction Operation and

design design maintenance

Figure 5.1 Opportunities for risk reduction

Appointment stage Management issues Design Construction Use Maintenance Decommission

Clear brief in place Competence and Method of analysis Site specific hazards Design records Maintenance Understand original
before resource of team retained and requirements design and
commencement updated outlined to client? modifications

Compatibility with Method of design Ground conditions Advice sought on Access for Adequate
others in team and existing services change of use maintenance information?

Conflict with Analysis model

statutory duty?

Adequate time? Co-operation Review and Societal risk Demolition of similar

measures checking appetite structures

Adequate fee? Co-ordination Construction/ Temporary works Asbestos and other

measures erection strategy toxic materials

Competent to do the Information required Maintenance Falsework

work? strategy

Special risks (lessons Information flow Special or innovative

from history?) procedures structure?

Form of Accidental loads

engagement?

Beyond codes?

Site wide issues

Procurement
Note
This table indicates one way an organisation could capture typical activities to reduce risk. It is intentionally incomplete; appropriate content may be selected
from elsewhere in this Report to suit the structure and activity concerned.

Figure 5.2 Prompts to consider in hazard elimination and risks reduction

indicative, as in practice opportunities will vary 5.4 How to do ‘risk assessment’

between projects.

Figure 5.2 shows an example of a list of prompts 5.4.1 General

which could be used in risk reduction. The
appropriate prompts will vary depending on the ‘Risk assessment’ is often used to mean both the
user and the stage in the construction process. process for managing hazards and their potential
The example does not show recommended consequences and for the document produced to
actions; it is for the engineer to determine what is record that process. ‘Risk management’ is a better
appropriate. description of the activity but since ‘risk assessment’

20 The Institution of Structural Engineers Risk in structural engineering

 How to manage risk 5.4

Identify hazards

Can
Yes
the hazard be Do it
eliminated?

No

Can
the hazard be Yes
substituted or Do it
isolated?

No

Can
risk (probability Yes Can Yes
or consequence) you do Do it
reasonably be this?
reduced?
No
No
Inform whoever
should do it

Is the Yes
residual risk Next hazard
acceptable?

No

Think again

Figure 5.3 Flowchart of a risk management process

is in common use, this Report uses the terms Each hazard, and then each remaining risk, should
interchangeably. It is about much more than be considered using the hierarchy in Section 5.3.
assessing risks and also includes eliminating hazards, How this is done will depend on the stage in the
reducing risks and communicating information about structure’s life cycle, and how much has already been
residual risks. It is not about production of pieces of decided. The more serious the risk, the more effort
paper. It is about understanding what might go should be put into managing it. If carefully chosen,
wrong, and how to prevent it, or at the very least to action to reduce one risk may reduce several others.
minimise the damage. On the other hand, care should be taken to avoid
introducing new hazards unless that reduces the
5.4.2 Process for risk management overall risk. While a good risk assessment at the
design stage will make the constructor’s risk
A flowchart for risk management is shown in assessments much easier, the nature of the project
Figure 5.3. This should, if possible, involve client, will have a major influence on the risk. For example,
designers and constructors. Each should be constructing a city centre deep basement on a
responsible for their own scope of work but should cramped site while retaining the original building
co-operate and exchange information. Structural façade will require very thorough risk assessment by
engineers should try to ensure that the client the constructor, however good the design is.
understands that early appointment of the
constructor and the extension of the designer’s 5.4.3 Documenting the risk assessment
contract into the construction phase will facilitate
communication and hence risk reduction. This is the There are two main reasons for documenting the
case for risks to health and safety and for risks to the work done. Firstly, to produce a risk register to
successful completion of the project. communicate the residual risks. Secondly, as an

The Institution of Structural Engineers Risk in structural engineering 21

5.4 How to manage risk

Ref no. Hazardous Hazard Measures taken to Measures not Information provided Date Action Date
activity eliminate hazard considered (or residual hazard(s)) issue required actioned
reasonably raised by
practicable
Construction Lack of Demonstration of
competence competence required in
tenders

Ground Contamination, Ground investigation

conditions slope or structure completed, results issued
instability

Existing services Danger to workers, Design based on service Full survey at all Service drawings
disruption to drawings depths prior to provided, CAT scan to
supply commencement be done before each
excavation
Temporary Instability
works

Various Work at height

Load deliveries Falling or swinging Provision made for lifting Detail design by
load fabricator should
include lifting provision

Working in a Asphyxiation, Structure designed for

confined space noise, inundation, construction without
etc. creation of confined
spaces
Placing concrete Dermatitis Some elements precast in
factory conditions

Erecting Falling from height Number of connections at Eliminate all

structural steel height minimised connections

Masonry Manual handling Block sizes limited

Note
This table indicates one way an organisation could communicate residual risk. It is intentionally incomplete; the risks relevant to the project need to be
identified by the project team. The risks listed here are only to illustrate the format.

Figure 5.4 Example format for a residual risk register

auditable trail for internal reference and to concerned, i.e. the auditable trail for design is fixed
demonstrate to the authorities, if necessary, that due when the design is complete.
process was followed.
Risks to health and safety may be included in the
Although the residual risk register and the auditable same register as risks to successful completion of the
trail are usually combined in a single document, it is project, or each could be in a separate register. On
useful to consider these two functions separately, to larger projects, registers can be divided by phase or
understand which information should be included. area; whichever is clearest and has least opportunity
for misunderstanding at interfaces.
The risk register is primarily required for
communication. It should describe non-obvious, A document on its own cannot reduce risks; only the
unusual or significant hazards so that the risks can be intelligent use of its conclusions can do that. The
managed and should list the risk reduction measures Industry Guidance for Designers5.1, published by
which require implementation by others. Too much ConstructionSkills and supported by the Institution of
data which is obvious will reduce its clarity. If a hazard Structural Engineers and HSE contains a section on
has been eliminated but could be reintroduced by hazard elimination and risk reduction, which includes as
later actions, e.g. the steelwork has been designed Section 2.7 some suggested headings for recording the
so that no connections are required at height, this process and outputs of a risk assessment. It does not
should be made clear. The risk register should be a suggest quantifying risks, either numerically or as high/
live document that will change and evolve during the medium/low, as a necessary part of a risk assessment.
design and construction. An example of a risk The Institution shares this view. It may be useful to draw
register is included as Figure 5.4. attention to the estimated probability or consequence,
particularly if severe or higher than usual for the hazard,
The auditable trail is primarily required as a record. It but there is no value (even if it could be done accurately)
should include all hazards, even those which are in scoring risks before and after risk reduction.
obvious or have been eliminated, and all risk
reduction measures already identified or Marking risk information onto drawings can be a
implemented. It remains live only for the activity good way to communicate this from designer to

22 The Institution of Structural Engineers Risk in structural engineering

 How to manage risk 5.5

constructor, but such information should be actions, errors and material failures appropriate to the
meaningful. There is no benefit simply listing hazards structure considered. To conduct the HAZOP
the constructor will be already aware of, or making effectively, it is a prerequisite that a ‘design’ already
bald statements to the effect that risks have been exists and that the team debating it are
assessed and reduced without saying what the knowledgeable about that design.
residual risks are.
A HAZOP may identify the potential for hazards such
5.4.4 What to do and not to do in risk assessment as explosions. If the risk cannot be eliminated, it
should be taken into account in the overall plant
Do layout. Building structures to resist explosions can be
– Demonstrate that risks can’t reasonably be reduced very expensive, and risks can be reduced simply by
further. careful siting so that the consequential damage is
– Think about the hazards and risks early in design. limited. The number of people killed in the explosion
– Design so that there is at least one safe way to and fire on the Piper Alpha oil platform in the North
erect the structure. If it is not obvious explain what Sea5.4, 5.5 (1988) would have been much less if the
provisions the design makes for construction. accommodation module had not been sited so close to
– Tell anyone involved in the work what the residual the main gas pipes, and in Texas City5.6 (2005) the 15
risks are. engineering staff would not have been killed if their
– Talk to the other parties (client, designers, main temporary offices had not been sited close to a major
contractor, subcontractors, fabricators, suppliers) – refinery plant.
try to minimise contractual barriers to
communication.
– Consider risks over the life of the structure.
– Record, for the owner/operator, how the design 5.6 Codes of practice
minimises lifetime risk.
– Use risk assessment as a process to understand
and record risks in that project. Codes of practice generally provide guidance to
– Put an amount of work into risk reduction which is designers. Many contain the stipulation that they
proportionate to the risk. should be used by qualified and experienced
engineers. In effect, the judgements that are common
Don’t to most designs have been taken by the authors of
– Focus on demonstrating that the risk is lower than the code, and the results set down in a manner that
it was initially (this only tests the starting position). can be applied in design. By using a code of
– Produce paper to tell constructors things they practice, the engineer implicitly accepts those
already know. judgements, in many cases without fully
– Treat risk assessment as a process to produce a understanding the basis for them, or the limits on
document. their application.
– Instruct others how to erect a structure.
– Carry out risk assessment just to protect against In particular, codes assume that the structures they
liability – do it to reduce risks. are applied to are ‘normal’ structures. Designers of
unusual structures need to understand this; for
example, clad, framed structures do not usually suffer
fatigue from wind induced oscillation, and thus most
5.5 HAZOP building codes do not address the phenomenon.
Slender and/or exposed structures such as chimneys,
masts or some sculptures may need design provisions
For structures containing hazardous materials, with beyond the code. In addition, every code of practice
complex performance demands or within an unusual makes assumptions about the acceptable level of
environment, a formal risk analysis using the safety5.7. These are set to meet the expectations of
principles of a hazard and operability (HAZOP) study society, which can vary from country to country
may be worth considering, either as written or with depending on the economic balance. The Eurocodes
modifications. This process is unlikely to be have therefore included Nationally Determined
applicable to ordinary structures, but an Parameters to enable each nation to set the safety
understanding of it may be useful to engineers who factors deemed appropriate for their locality and society.
are developing their understanding of risk.
In the UK, codes of practice are not mandatory but
HAZOP was developed by Trevor Kletz5.3 in 1983 as a those listed in the Approved Documents that support
tool for the chemical process industry. The process the Building Regulations5.8 are deemed to satisfy the
uses a systematic approach to identify any possible regulations; the designer may choose another method
deviations from the design intent, and any consequent but is responsible for showing that the design is safe. In
hazards. It is of great value, particularly for assessing the USA, the courts tend towards a similar
any complex system where a mixture of engineering interpretation. In some countries, including Spain and
disciplines has been deployed and where plant failure is Italy, non-compliance with codes of practice is in itself a
a possibility. It was not developed for use in structural criminal offence, irrespective of whether failure occurs.
engineering, but the principles on which it is based are
applicable to any engineered system.

As originally conceived, HAZOP was based on key 5.7 Risk management framework
words that reflected the origins of the approach in
process industries, but its use in those industries
(including nuclear power) has expanded to review all For ongoing operations it can be beneficial to have a
types of engineering systems. Appropriate key words risk framework that identifies all potential risks,
should be selected to represent potential loads, obtains data on their likelihood and is regularly

The Institution of Structural Engineers Risk in structural engineering 23

5.8 How to manage risk

reviewed to detect trends and changes in trends. An the particular project and may highlight to the client
example of this is the UK Rail Safety and Standards and designer where gaps in experience/knowledge
Board’s (RSSB) Safety Risk Model5.9. This provides appear and where those gaps may be filled by
risk data covering the management of railway risks and specialists.
includes details of near misses, which are particularly
important in assessing the risk of low probability/high
consequence events. It also uses the concept of
equivalent fatalities to address the risk of ‘minor’ or 5.9 QA and change control
‘major’ injury. There is no similar framework for
structural engineering risk, but the UK’s CROSS
(Confidential Reporting on Structural Safety)5.10 Systems for management of work, otherwise known
scheme captures and collates information on matters as quality assurance (QA), are a valuable way to
of concern to structural engineers and permits reduce risks due to misunderstandings or errors.
monitoring of trends. CROSS was established by They rely on a careful definition of good practice,
SCOSS (Standing Committee on Structural Safety, intelligent application and continuous improvement. It
www.structural-safety.org) in 2005 and publishes is important that users recognise that QA does not
quarterly newsletters for structural engineers. replace engineering judgement, but only provides a
framework to support it. If the rules are too complex
or inflexible they will tend to be followed blindly, with a
risk that the wider issues are overlooked. Many QA
5.8 The importance of competence systems focus on detail rather than ensuring that
competent people and systems are used and, as a
result, can become bureaucratic and ineffective.
Risk management relies on the judgement of the
engineers involved; it cannot be carried out by rote. It Control of changes to the design is a key role for QA.
is therefore important that engineers and the All changes should be authorised by the person in
organisations they work for are competent. In the UK, charge of the design. Every proposed change should
there is a legal requirement for competence under the be checked against the design intent, to ensure that
CDM Regulations. Guidance was published in the change does not invalidate design assumptions
20065.11 and remains good advice, although the or other parts of the design. If a change remains
regulations were revised in 20075.12. within the design intent, and is executed carefully, it
should carry little risk. Changes that modify the
Competence requires an understanding of what has original intent require more detailed scrutiny. There
to be done and the experience and ability to do it. remains a need for judgement, as serious accidents,
Apart from formal training and education, such as the Hyatt Regency Hotel5.14 (1981, see
competence is usually obtained by on the job Appendix A.15), have occurred due to design changes
training; working under the guidance of senior which were far more significant than were realised by
engineers. All structural engineers should aspire to those who made them.
raise their personal level of competence, and
members of the Institution have a professional
responsibility to develop their skills through
Continuing Professional Development (CPD)5.13. 5.10 Independent review
Management of structural engineering risk typically
requires experience of design and construction. The
ability of designers to reduce risk is frequently limited by Independent ‘third party’ review of designs is often
a lack of construction experience. Learning constantly used where risks are high, and could be applied
from what has gone wrong elsewhere is an essential beneficially on many projects. In the UK, it is standard
risk reduction exercise. practice for railway bridges, for highway structures
(Category 1, 2 and 3 independent checks) and for
In any organisation there is bound to be a gradation nuclear facilities. A Guidance Note5.15 has been
of skill and experience and the risks inherent in that issued by SCOSS describing some of the features and
diversity need to be managed. It is therefore benefits of one approach to independent review.
important that less experienced engineers are
allowed to develop their skills and experience in risk Even when independent review is not a legal or
management, subject to review by senior engineers. contractual requirement, a review of the overall
This applies not just in the design office but on site; design by a third pair of eyes, independent of the
safe management of both temporary works and the original designer and verifier, should be standard
stability of part finished structures requires both practice for all designs
structural engineering competence and a willingness
to take firm action in the face of commercial
pressures if risks are not recognised.
5.11 Conclusions and recommendations
Many clients formally review the competence of the
organisations they employ or propose to employ on
construction projects. This minimises their risks and Project specific and unusual hazards should be
in some countries, including the UK, is a legal identified early and kept under review. A hierarchy of
requirement. It is usual to measure the qualifications risk reduction measures should be applied during
and relevant experience of individual key members of both design and construction. It is more important to
staff, as well as the corporate competence of the carry out a thorough and proportionate risk
organisation, as demonstrated by its staff, structure assessment than to document it, but communication
and performance. Self assessments by designers and of residual risk is essential. Drawings are a good
suppliers enable informed decisions by clients. These medium to communicate construction risk. These
assessments should include resource availability for activities should be focussed on what is useful to

24 The Institution of Structural Engineers Risk in structural engineering

 How to manage risk 5.12

reduce real risks and not on producing paperwork; 5.13 Engineering Council. UK-SPEC [UK Standard for
listing trivial or well understood risks is neither Professional Engineering Competence]. Available at:
necessary nor useful. http://www.engc.org.uk/ukspec [Accessed: 18 February
2013]
Demonstrating that risk has been reduced compared
to the original design only illustrates that the concept 5.14 Marshall, R.D. et al. Investigation into the Kansas City
was flawed; the point is to end up with risk reduced Hyatt Regency walkway collapse. NBS Building Science
to the lowest reasonable level. Codes of practice are Series 143. Washington, DC.: NBS, 1982
a useful guide, within their scope, but competence
and the ability to apply engineering judgement are 5.15 Standing Committee on Structural Safety. Independent
essential. review through peer assist. SCOSS topic paper SC/09/
034. Available at: http://www.structural-safety.org/
topicpapers [Accessed: 18 February 2013]

5.12 References

5.1 CITB-ConstructionSkills. Industry Guidance for

Designers, King’s Lynn: CITB-ConstructionSkills, 2007.
Available at: http://www.cskills.org/uploads/CDM_
Designers4web_07_tcm17-4643.pdf [Accessed:
18 February 2013]

5.2 Carpenter, J. ‘Risk management with ERIC’. The

Structural Engineer, 88(7), 7 April 2010, pp20-21

5.3 Kletz, T.A. HAZOP & HAZAN: identifying and assessing

process industry hazards. 4th ed. London: Taylor &
Francis, 1999

5.4 Crawley, F. ed. Piper Alpha: lessons for life cycle safety
management. Rugby: IChemE, 1990

5.5 Lord Cullen. The Public Inquiry into the Piper Alpha
disaster. London: HMSO, 1990

5.6 Fatal Accident Investigation Report: Isomerization Unit

Explosion Final Report. Available at: http://www.bp.
com/liveassets/bp_internet/us/bp_us_english/STAGING/
local_assets/downloads/t/final_report.pdf [Accessed:
18 February 2013]

5.7 Standing Committee on Structural Safety. The

Assumptions behind the Eurocodes. SCOSS topic
paper, Nov 2009. Available at: http://www.structural-
safety.org/topicpapers [Accessed: 18 February 2013]

5.8 The Building Regulations 2010 (SI 2010/2214).

Available at: http://www.legislation.gov.uk/uksi/2010/
2214/pdfs/uksi_20102214_en.pdf [Accessed: 7 March
2013] [Note that similar but separate regulations apply
in Scotland and in Northern Ireland, see http://www.
legislation.gov.uk/all?title=building%20regulations]

5.9 Rail Safety and Standards Board. Safety Risk Model.

Available at: http://www.rssb.co.uk/srmodel.asp
[Accessed: 18 February 2013]

5.10 Confidential Reporting on Structural Safety [CROSS]

website. Available at: http://www.structural-safety.org
[Accessed: 18 February 2013]

5.11 Carpenter, J. Developing guidelines for the selection of

designers and contractors under the Construction
(Design and Management) Regulations 1994, HSE
Research Report 422. Sudbury: HSE Books, 2006.
Available at: http://www.hse.gov.uk/research/rrpdf/
rr422.pdf [Accessed: 18 February 2013]

5.12 The Construction (Design and Management)

Regulations 2007 (SI 2007/320). Available at:
http://www.opsi.gov.uk/si/si2007/20070320.htm
[Accessed: 19 February 2013]

The Institution of Structural Engineers Risk in structural engineering 25

6 Statistical and probabilistic methods

6.1 Introduction experienced individuals. In advanced structural

engineering, ‘reliability’ can be assigned a numerical
value6.2. BS EN 19906.3, the head Eurocode, offers a
This chapter provides an overview of the numerical methodology for specialist calculation of an appropriate
background to risk management. It is mainly relevant load factor taking account of views on the accuracy of
to structural stability rather than hazards on analysis, confidence on material properties and so on.
construction sites. It is written for practising structural
engineers, not risk specialists, and is not intended to
be academically rigorous. A bibliography is provided
for further study. 6.3 Quantifying probability

There are various ways of quantifying probability. In

6.2 Background everyday language the chances of winning a lottery
might be considered as 1 in 14 million per ticket
purchased, or the probability of dying from smoking
Day to day structural engineering does not generally as 1 in 200 per year. In engineering practice, rational
require direct application of statistical methods. methods of defining ‘safety margin’ or ‘probability of
Indirectly, however, statistics have a significant failure’ can be explored using numerical methods
influence, being used extensively in assessing and and, in some branches of engineering, such methods
describing material strengths and in describing the are used quite widely. If enough similar items are in
likelihood of random loadings such as wind. For use, the proportion that fail can be counted and used
designs that meet the requirements of a code of to calculate the probability of others failing in the
practice, the safety factors reduce the probability of future. Thus, for a complex electrical system, with
failure to a low enough level to take account of the known failure probabilities for the separate items
uncertainties in design, loading and material quality. within that system, it is mathematically possible to
An understanding of statistics and probabilities can define the overall level of reliability or, looked at the
be useful in decision making and in assessing the other way around, the overall probability of failure.
capability of existing buildings where code shortfalls This can be compared with what might be
have been identified. considered an acceptable risk. This would also need
an understanding of the way components interact,
More widely, probabilistic approaches can provide such as the possibility that one failure could trigger
useful insights for comparing risks, providing another, or that one event could cause multiple
perspective to help decide whether a risk is tolerable, failures.
to decide which risk requires most attention or to
support a decision that the risk has been reduced to To apply this process to structures, there would need
an acceptable level. to be some basis for counting actual failures. This is
not really available, certainly for the whole structure.
Since the magnitude of many loads on a structure is The reliability of electronic components can be
fundamentally uncertain, they can only rationally be calculated easily, as thousands are made to the same
defined in terms of probabilities. An example is the 1 design. Few structures are identical and as explained
in 50 year wind, defined as the wind speed with a 1 elsewhere in this Report, many of the causes of
in 50 probability of being exceeded at least once failure are independent of the design code used or
during a year. This gives a 64% chance that it will be the precise details of the structure. One area where
exceeded at least once in 50 years. Low probability failure statistics are used to good effect is in
events do happen and this wind speed could be preparedness for emergency response. In most parts
exceeded more than once in the period or even in the of the world, buildings follow a common form, e.g. in
first year. For structures where very high reliability is the UK, the brick built three bedroom semi-detached
sought, a 1 in 10000 year event might be used6.1. house. In an area of high seismicity, there will be a
There is a debate about the use of probability as the historical understanding of the response of the local
sole way to deal with such low frequency natural type of houses or bridges to earthquake loading of
events, compared to the resilience approach described certain intensities and it is possible to make an
in Section 3.10. estimate as to how many might be damaged in a
particular earthquake and hence decide what
The use of numerical assessment can be deceptive, preparedness could be undertaken. This approach
suggesting a degree of precision that rarely exists. A becomes more effective as more data is available,
probability might be calculated as 1 in 10000 per such as through improved seismology and satellite
year (often written as 10 4 per year), but due to the based damage surveys.
many assumptions which have to be made in the
assessment it might in practice lie between 10 2 and Calculated probabilities will only be correct if the data
10 6. Hence these methods are somewhat used to generate them is correct. For example, it may
approximate and should not be allowed to dominate be assumed that concrete meets the specified
any risk assessment; the techniques are best used in characteristic strength; if, as often happens, it is over
conjunction with deterministic methods. Moreover, strength, the probability of failure may be lower
because this field is so specialised, the preparation of (provided the amount of reinforcement is adequate).
probabilistic numerical arguments is best left to It is usual to calculate the probabilities using what are

26 The Institution of Structural Engineers Risk in structural engineering

 Statistical and probabilistic methods 6.4

known as ‘conservative’ values, that is, the value is and mortar used. This approach recognises the overall
chosen cautiously, so that the final probability of lower probability of failure if workmanship is controlled
failure is probably an overestimate. It is often argued more closely. Partial factors are also varied for different
that conservative values should be used where there load combinations (e.g. dead þ live þ wind) to reflect
is uncertainty, but this approach requires care. Risk the lower overall probability of that combination of
management often involves compromises, comparing circumstances arising. It is important to be aware that
one option against another to see which has the some uncertainties, such as the accuracy of our
lowest overall risk6.4, but inconsistent levels of models of structural behaviour, do not have a specific
conservatism could overestimate one side of the partial safety factor but are included within other
balance, and so skew the judgement. factors. This means that even if the load is known
exactly, a partial factor for load of 1.0 may be
inappropriate, as the factor also covers other, unstated
aspects which are still uncertain.
6.4 Safety factors

To avoid the need for engineers to apply probability 6.5 Low probability events
theory in routine design, structural design generally
includes a safety factor or load factor (usually built up
from partial factors) which ensures that the probability For events with low probability but high
of overall failure is low enough. The value of the factor consequences, the reliability provided by standard
was traditionally subjective, based on collective design codes may not be adequate. This becomes
historical experience, but there is now an aspiration to more relevant as the population of the world grows
derive the factors statistically or by reliability theory. and the number of people living in areas vulnerable to
A structure with design factors lower than in the infrequent but severe natural hazards such as floods
design code is not necessarily unsafe, but it should be and earthquakes increases. Coupled with the wider
expected to have a higher probability of failure than development of an engineered infrastructure, there is
one that meets the code. Excluding gross error, a growing belief among the public that the human
structural failure would not normally occur unless the consequences of such disasters are avoidable.
combined probability of adverse variation in applicable Nevertheless, communities still have to cope with the
loading, loading configuration, material quality, aftermath of severe wind, severe temperatures and
workmanship, etc., all become coincidentally too high. severe flooding even in countries with a
well-established infrastructure such as the UK and
Structures meeting the code requirements, therefore, USA. A practical question engineers have to address
have an acceptably low probability of failure. If very is to determine the likelihood of such events and then
high reliability is sought, this can be effected by decide what resources can be afforded to defend
increasing the load factor, using reliability theory to against them. Techniques such as quantitative risk
obtain a specific increase. Conversely, if an increased assessment (QRA) are available to predict the
probability of failure is acceptable, perhaps because a (numerical) magnitudes of rare natural events and to
structure need only have limited functionality after the predict the likelihood of process plant failures.
event, then the required load factor may be reduced.
For example, a building under construction would be An early use of QRA6.7 was in the study carried out in
unoccupied and construction work would usually 1978 to assess the risk that the chemical plant on
stop in very strong winds, so the design wind loading Canvey Island posed to London6.8. Major accidents to
during construction may be based on a two year such plants do happen, such as the failure at
recurrence period rather than 50 years. Flixborough6.9 (1974) or at Buncefield6.10 (2005). How
can engineers decide which modes of failure are
‘Time at risk’ may be considered as a reason to possible, what the probabilities of those failures are and
reduce the factor of safety. This may be valid for hence the risk they pose in terms of their location
transient risks or those that could occur during only a relative to populated areas? These risks can be
small fraction of the life of the structure and where it expressed in numerical terms. In areas where the
might be disproportionate to use the usual factor. population are at risk from flood, an event with a
When designing for a specific situation that only lasts statistical probability of occurrence, one way of
a short time, such as the construction phase example assessing the likelihood of damage, or the required
in the previous paragraph, this may not be valid, and height of a flood protection system, is to assess the
each case should be taken on its merits. For the risks using statistics and numerical values against a
people involved, construction is usually a full time target acceptance value. In London, a quantitative
activity, on one site after another, so increasing their flood risk assessment is required to review the
risk would be unjustified. effectiveness of the Thames Barrier6.11. The
consequences of the floods in New Orleans following
Confidence in material properties clearly affects the Hurricane Katrina6.12, 6.13 (2005) illustrate the need.
selected value of a load factor. Material testing is
bound to show a scatter of results and this is In the nuclear industry, when assessing the safety of
managed by using such concepts as the 95% facilities, a combination of deterministic (designing for
confidence level, which typically defines the specified events) and probabilistic assessment (the
characteristic strength. probability of those events not occurring) is used.
Neither method is used exclusively, since both give
In some codes, the partial factors are varied explicitly insights into the overall safety of the plant.
to take account of known factors affecting the Probabilistic methods are often used to define the
probabilities. For example, in UK masonry design deterministic events. The accepted ‘safe’ target in the
codes (both BS 56286.5 and BS EN 19966.6 with UK UK is that the probability of significant harm to the
National Annex) the partial factor for materials depends public from radioactivity should have a probability of
upon the workmanship and quality control of blocks 10 7 per year or lower. In practice, events as rare as

The Institution of Structural Engineers Risk in structural engineering 27

6.6 Statistical and probabilistic methods

this are very difficult to predict so the design basis for to design for the worst possible case. Considering
natural hazards such as earthquakes is an event that, the potential consequences will allow the acceptable
by a conservative assessment, has a 10 4 per annum probability of occurrence to be calculated; if the
probability of exceedance6.1. This may be compared to probability of the worst possible case is lower than
the typical UK loads of 1 in 50 years for building this, the design may be based on the limiting
structures and 1 in 120 years for bridges. Use of the acceptable risk. Numerical probability was used in the
10 4 event with a conservative approach to design design of the river barriers for the London Eye6.17.
ensures that the combined probability of the Given the client’s wish to site the observation wheel on
earthquake, leading to structural failure, leading to the edge of the River Thames, any additional risk had to
release of significant radioactivity, is in the region of be reduced to a level consistent with a more
10 7 per year. The 10 4 earthquake for the UK is a conventional site. It would be geometrically possible for
significant design loading. Design also considers a laden vessel to hit the Eye’s capsules in certain river
corresponding extremes of environmental loading such states, if this was not prevented by the pontoon and its
as wind and temperature. Similar approaches are used moorings. The design question was ‘what impact
in probabilistic studies of risks to offshore installations, energy should the pontoon be designed against?’ The
railways or from ship impact on bridges6.14. designers identified one specific vessel that operates
occasionally and could, if fully loaded and in
Rather than consider a single earthquake return combination with specific tide and flood conditions and
period, with an estimated probability of structural other specific factors about the vessel’s approach,
failure in that earthquake, a more advanced approach produce a particularly high impact load. A more likely
would be to consider the probability of a range of event would be for one of the very much lighter but very
earthquake intensities, to calculate the failure much more frequent tourist vessels to hit the pontoon
probability for each and to combine these. This could as they manoeuvre in front of the Eye. Modelling the
be done by numerical integration of the earthquake impact and considering the probability of each event
hazard curve and the structural fragility curve. The along with the cost of designing and building the
difficulty with this approach is that it requires actual pontoon to resist it allowed the designers to reach a
fragility data for the type of structure concerned, judgement on which approach was reasonably
when subject to the relevant loading, and such data practicable.
does not exist for unusual structures or areas with
infrequent earthquakes. Designing routine buildings for a lower wind loading
during construction is rational. But if the
When acceptable risks are expressed in terms of a consequences resulting from failure in a slightly larger
failure probability per year, care has to be exercised wind were really intolerable it pays to think again, for
when exposure to the risk is only for a short period. it might only cost a marginal amount more to reduce
As the time interval of exposure is short, it might the probability of failure. Likewise, it is irrational to
appear that the probability of failure ‘at that instant’ is design every structure for terrorist attack. In reality, it
very low. But if there are many such exposures, the is necessary to judge the probability of the event and
probability that a failure will occur at some stage is define some design standards accordingly. For most
much increased. Appendix A.12 discusses the buildings, the likelihood of terrorist attack is low, and
probability of failure in the context of the road/rail therefore the associated risk is tolerably low when the
accident at Great Heck6.15 in Yorkshire, UK (2001). cost in potential loss of life and injuries is considered.
The risk increases when the target has a higher
probability of attack and especially where the
consequences are of harm to large numbers of
6.6 Application people. In effect, the classes of robustness6.18 used in
routine design are standardised judgements taking
account of the consequences of failure. Thus,
The approaches described above may seem esoteric structures are divided up into groups with more care
and distant from the real world but have many being taken on those structures where the
practical applications. Firstly, consciously looking at consequences of failure in terms of loss of life are
any structure in such terms allows rational decisions greatest. Guidance on design against accidental
to be made on where to spend money to improve loadings, including specific reliability assessment for
safety. Thus, putting the whole problem in terms of high consequence events, is given in Part 1-7 of
the uncertainties, the effect of changing a particular BS EN 1991, Eurocode 16.19.
parameter can be investigated to understand how
much it costs and how much ‘safety’ it buys or what Although the nuclear, chemical and aeronautical
risk reduction is achieved. Effort should always be industries continue to apply probabilistic analyses to
focussed on the most significant risks. Similarly, the guide their projects and choices, such analysis for
cost of designing a facility against a return period of normal structural projects remains unusual. This may
10 4 flood as compared to 10 2 can be investigated. be at least partly due to the difficulty of maintaining
In Holland, the consequences of spring floods on the databases of component reliabilities. The nearest
Rhine overtopping the dykes have led to design risk of approach is the use of characteristic strength to
1 in 1250 years, while the sea defences in the west of manage the probability of faults in metal castings,
the country are designed for a 10 4 event6.16. This cavities in concrete piles or under-strength concrete.
does not mean they will not fail in 10 000 years;
improbable events can happen. It is also likely that
forecasting such a rare event on the basis of 100 years
of records will not be accurate. Estimates of extreme 6.7 Assessment of existing structures
values are updated as knowledge improves, and as the
climate changes.
Structural engineers are frequently faced with the
Probabilistic techniques can, where appropriate, need to appraise structures that do not comply with
provide a rational way to show that it is not necessary ‘modern standards’. This does not necessarily mean

28 The Institution of Structural Engineers Risk in structural engineering

 Statistical and probabilistic methods 6.8

they are unsafe and it may not be in the best reinforced and unreinforced masonry structures.
interests of society to spend a disproportionate London: BSI, 1996 [Incorporating corrigenda February
amount of money ‘strengthening’ them. Many older 2006 and July 2009 and UK National Annex]
structures were designed with a much larger factor of
safety than today’s structures because of lack of 6.7 Det Norske Veritas. A Comparison of accident
knowledge about material performance over time and experience with Quantitative Risk Assessment (QRA)
the inability to analyse structures as rigorously as is methodology, Contract Research Report 293/2000.
possible today. Depending on where the ‘real’ Sudbury: HSE Books, 2000. Available at: http://www.
performance lies, the actual factor of safety may be hse.gov.uk/research/crr_pdf/2000/crr00293.pdf
very large, or quite small. With a knowledge of how [Accessed: 18 February 2013]
safety is defined in terms of failure probability, it is
possible to assess a structure taking into account the 6.8 Health and Safety Executive. An Investigation of the
actual uncertainties to judge the failure probability. For potential hazards from operations in the Canvey Island/
example, the actual loading may be known accurately Thurrock area. London: HMSO, 1978
and there may be evidence of strength being much
better than assumed in design. In engineering terms it 6.9 The Flixborough disaster: report of the Court of Inquiry.
is legitimate, in this judgement, to take benefit from London: HMSO, 1975
other structural qualities such as redundancy and
ductility, but in countries where design codes are 6.10 Explosion Mechanism Advisory Group report. London:
mandatory rather than advisory, the legality of such Buncefield Major Incident Investigation Board, 2007.
approaches should be checked. Available at: http://www.buncefieldinvestigation.gov.uk/
reports/buncefieldagr.pdf [Accessed: 18 February 2013]
Further information is also available in ISO 13822:
20106.20, Assessment of existing structures, and in the 6.11 Dawson, R.J. et al. ‘Quantified analysis of the
Institution’s report, Appraisal of existing structures6.21. probability of flooding in the Thames Estuary under
Imaginable worst-case sea level rise scenarios’. Water
Resources Development, 21(4), December 2005,
pp577–591. Available at: http://www.hm-treasury.gov.
6.8 Conclusions and recommendations uk/d/atlantis-floodmodellingpaper.pdf [Accessed:
18 February 2013]

Normal design codes contain design methods and 6.12 Select Bipartisan Committee to Investigate the
embedded load factors which experience has shown Preparation for and Response to Hurricane Katrina. A
will usually produce safe structures. Design of Failure of initiative: final report of the Select Bipartisan
ordinary structures subject to normal loading should Committee to Investigate the Preparation for and
be carried out in accordance with design codes and Response to Hurricane Katrina. Congressional Report
there is no benefit in using probabilistic methods to 109-377. Washington, DC: USGPO, 2006. Available
manage risk in such designs. at: http://www.gpoaccess.gov/serialset/creports/katrina.
html [Accessed: 19 February 2013]
Probabilistic methods can be valuable for gaining a
deeper understanding of the risk resulting from 6.13 American Society of Civil Engineers. What went wrong
complex structures, containment of hazardous and why: the New Orleans Hurricane protection
materials or unusual loading. They can also be useful system. Reston, Va.: ASCE, 2007
in assessment of existing structures.
6.14 Duckett, W. ‘Risk analysis and the acceptable
probability of failure’. The Structural Engineer, 83(15),
2 August 2005, pp25-26
6.9 References
6.15 Health and Safety Executive. The track obstruction by a
road vehicle and subsequent train collisions at Great
6.1 Health & Safety Executive. Safety assessment principles Heck 28 February 2001. Sudbury: HSE Books, 2002.
for nuclear facilities. 2006 Edition, Revision 1, Available at: http://www.rail-reg.gov.uk/upload/pdf/
paragraph 514. Available at: www.hse.gov.uk/nuclear/ incident-greatheckfinal-optim.pdf [Accessed:
saps/saps2006.pdf [Accessed: 18 February 2013] 19 February 2013]

6.2 ISO 2394:1998: General principles on reliability for 6.16 Hoekstra A.Y. and De Kok, J-L, ‘Adapting to climate
structures. Geneva: ISO, 1998 change: a comparison of two strategies for dike
heightening’. Natural Hazards, 47, 2008, pp217–228.
6.3 BS EN 1990:2002+A1:2005: Eurocode – Basis of Available at: http://doc.utwente.nl/59991/1/
structural design. London: BSI, 2010 [Incorporating Hoekstra08adapting.pdf [Accessed: 19 February 2013]
corrigenda December 2008 and April 2010]
6.17 Beckett, T. ‘The British Airways London Eye. Part 5:
6.4 BOMEL Ltd. The global perspective in addressing Pier and impact protection system’. The Structural
construction risks. Research Report 458. Sudbury: Engineer, 79(2), 16 January 2001, pp34-35
HSE Books, 2006. Available at: http://www.hse.gov.uk/
research/rrpdf/rr458.pdf [Accessed: 26 February 6.18 Institution of Structural Engineers. Practical guide to
2013] structural robustness and disproportionate collapse in
buildings. London: IStructE, 2010
6.5 BS 5628-1:2005: Code of practice for the structural
use of unreinforced masonry. London: BSI, 2005 6.19 BS EN 1991-1-7:2006: Eurocode 1: Actions on
structures – Part 1-7: General actions – Accidental
6.6 BS EN 1996-1-1:2005: Eurocode 6 – Design of actions. London: BSI, 2010 [incorporating
masonry structures – Part 1-1: General rules for corrigendum February 2010]

The Institution of Structural Engineers Risk in structural engineering 29

6.10 Statistical and probabilistic methods

6.20 ISO 13822:2010: Bases for design of structures –

Assessment of existing structures. Geneva: ISO, 2010

6.21 Institution of Structural Engineers. Appraisal of existing

structures. 3rd ed. London: IStructE, 2010

6.10 Bibliography

Blockley, D.I. The nature of structural design and safety.

Chichester: Ellis Horwood, 1980

CIRIA. Rationalisation of safety and serviceability factors in

structural codes. CIRIA Report 63. London: CIRIA, 1972

Diamantidis, D. ed. Probabilistic assessment of existing

structures: JCCS report. Available at: http://www.rilem.org/gene/
main.php?base=500219&id_publication=96 [Accessed:
18 February]

Kletz, T.A. ‘Process industry safety’. In Blockley, D I. ed.

Engineering safety. London: McGraw-Hill, 1992, pp347-368

30 The Institution of Structural Engineers Risk in structural engineering

7 Risk in design

7.1 Introduction harm occurs. These are not the only liabilities;
contractual claims for delay, additional costs or failure
of the structure to perform satisfactorily can be large.
7.1.1 Overview The costs of changing the design or of defending
claims for damages can be out of proportion to the
This chapter discusses what should be done at the cost of the work. Risk management should not be
design stage of a project to manage risk over the life seen as a cost but as an essential activity to control
of the structure. Many of the decisions taken at this costs.
stage can have a major effect on later risks. This is
not limited to structural design, nor to the scope of 7.1.4 What can go wrong?
the consulting engineer. Topics such as procurement,
programme and construction strategy are relevant to There are many opportunities for things to go wrong
minimising risk. during design, even when both individuals and their
organisations are fully competent. The majority of
7.1.2 What does the design stage include? structures are unique and large amounts of data are
both consulted and produced. The range of
Design is a key activity for structural engineers and is knowledge individual engineers need in order to
much wider than just the methodical process of identify errors is wide. While the technical press often
calculation and drawing production to meet code reports failures, the root causes are not usually
requirements. Designers of both permanent and known until later, when the failure is no longer
temporary works have an obligation to make sure newsworthy. Typical errors include failure to
that their design meets the functional demands of the understand the project requirements, an important
project brief, performs adequately and will be safe to feature being overlooked, mistakes in calculation or
construct, operate and eventually demolish. drawing, an imperfect understanding of the loads or
load path and imperfect technical understanding (e.g.
Design, therefore, means any part of the whole of concrete durability or dynamics). Failures also
process of producing a structure through from occur as a result of poor communication, either with
concept to completion, except the physical process other disciplines such as services engineers or
of construction. Most structural engineers are architects or between the designer and the (yet to be
involved in this in some way, even those not calling appointed) constructor.
themselves designers. Preferably, design should
involve clients and construction staff. The process Given the complexity of the task and potential
often requires interaction with other disciplines and consequences, the structural engineering profession
encompasses concept, functionality, evolution of has developed a culture of independent checks
structural form, calculations, drawings and within the design office. In recent years, some people
specifications, and procurement. Design is both a have thought this unnecessary or commercially
theoretical skill and a practical skill. unjustified. This is unfortunate; modern codes,
contractual models and computers have introduced
The design stage involves looking ahead to more complexity and more opportunity for error while
construction and use to see what can be done to reducing the margins in design. Checking remains an
reduce future risk. This may take the form of essential part of design.
planning, such as designing for a construction
sequence that provides weather protection as soon 7.1.5 Managing uncertainty
as possible, carrying out good soil investigation well
in advance, or choosing a particular design solution A key aspect of design ‘risk’ is to take the
to reduce risk. For example, piling can eliminate opportunity to minimise the effects of the
excavation in unstable ground. Careful tendering and uncertainties inherent in the design and construction
selection of contractors can reduce many risks by process. For example, delay can occur due to the
ensuring adequate competence. Restrictions on unavailability of materials or application of new
subcontracting can also reduce the number of technology. While these risks are not created by the
interfaces. structural engineer, there is an opportunity to
eliminate them by proper consideration during
The initial stages of the design process are the best design.
time to take strategic action to manage risk; if this
opportunity is missed, the deficit cannot be made up Work in the ground is fraught with risk and the term
during construction and use. ‘unforeseen ground conditions’ is familiar to every
experienced engineer. This risk can be minimised by
7.1.3 The consequences of design stage errors site investigation and by choosing a foundation
solution which reduces the effect of any remaining
If risks are not properly managed at the design stage, unknowns.
the worst case would be that people would be
harmed. That could result in both criminal and civil A useful tool is to test the sensitivity of the design
liabilities extending not only to those who ‘caused’ by varying the parameters over the range of any
the harm but those who failed to prevent it. Criminal uncertainty. Structures may be sensitive to
liability can also be incurred if legally defined variations in load, errors in construction geometry,
processes for designers are not followed, even if no defects in materials, and many other changes. The

The Institution of Structural Engineers Risk in structural engineering 31

7.2 Risk in design

key issue is that sensitive configurations are often the structure will be analysed, designed and detailed
fixed at the concept design stage, and should to achieve the overall design objectives. It should
therefore be identified and eliminated or reduced at identify a structural form that is functional in-service
that stage. and capable of being built without undue site risk.
The document should be agreed by relevant internal
Questions to ask might include: and external parties and periodically reviewed to
– Is the structure so sensitive to the prediction of confirm that it remains relevant and that the design
wind speed that a minor change will render the still follows it.
design unsafe?
– Is the deformation or alignment of the structure The design work should be executed in stages, with
really critical? interim checks to make sure no serious mistake is
– Can the dynamics be predicted with confidence? carried forwards7.2. If there is a major analysis on which
– Will any minor change in any of the design all subsequent member design is based, then it is
assumptions (such as tolerances) render the common sense to check that before handing it out for
structure weak, unstable or not functional? detail design.

Many risks are traditionally avoided by making the Every analysis, and especially every complex
structure insensitive; for example, by using ductile computer analysis, should be fit for purpose. Any
steel and proper member and connection design to computer analysis should have some hand checks to
ensure ductile behaviour so that the mode of failure in make sure its predictions are of the right magnitude.
overload is controlled. A general assumption is ‘the Computers can produce very precise output, but
stronger the better’. In certain circumstances, precision is not the same as accuracy. The predicted
however, such as impact, blast protection or in behaviour of the computer model is only relevant if
seismic design, structural performance is based on the model is a good representation of the structure.
controlled failure. In such circumstances, The Institution has published a report on the use of
over-strength in the wrong place can be detrimental. computers7.3.

An example of sensitivity is a lightweight temporary All design calculations should be self-checked by the
large television screen that had an apparently sound designer and then checked separately within the
margin of 1.5 against overturning at the low wind designer’s organisation. Even where they exist, it is
speed thought appropriate for design. It was not acceptable to rely on statutory external controls
considered ‘adequately safe’, but failed or independent reviews as the primary check.
catastrophically7.1. In the investigation it was observed
that a wind speed increase from 15m/s to 17m/s Ideally there should be concept checks, approximate
increased the wind force by 1.28 [¼(17/15)2], removing manual checks (using simple formulae such as
most of the margin. ‘WL/8’) targeted to ensure freedom from gross error,
followed by standard checks of structural elements.
Another approach to uncertainty is to choose design The systems that provide global stability should be
values on the safe or ‘conservative’ side. Care needs clearly identified.
to be taken to establish what is conservative with
properties such as the co-efficient of friction. If sliding Staff should be trained to set out their work so that it
is to be prevented, it is conservative to choose a low is obvious where all the data comes from, to assist
value, but if sliding is beneficial (such as when both checking and any subsequent changes.
dragging a load), a high value should be chosen. Calculations should be set out so that the output to
Similar issues occur with stiffness, depending on the drawings is clear.
whether deflection or strain controlled load is of
interest. If in doubt, a ‘best estimate’ value should be Detailed methodical checks on all drawings and
used and then varied in each direction to understand Building Information Models should be supplemented
the effect. with a look over the details by an experienced eye to
see if they look right. These should be suitable for
7.1.6 Project risk awareness their purpose, i.e. to communicate the design to the
constructor. At this stage, a check should be made
Awareness of the specific risks on a project can do that sensible precautions have been taken to assure
much to help eliminate or reduce them, or to ensure structural robustness (see Section 7.6). In developing
they are allocated to the right person or organisation. the structural scheme, designers should have regard
This is not a matter of shifting blame, more of to the future difficulties of inspection or the potential
ensuring that action is taken when required and for lack of durability. A designer’s professional duty is
providing a focus on what is important. It is a sensible to assure best value for money (not necessarily the
precaution at the beginning of a contract for all least initial cost).
parties to recognise where the risks lie and respond
accordingly. For example, if the programme for the Design office management should assure that the
design of a steel framed building is critical there is mechanics of the design process are carried out
absolutely no point in shaving off a tonne of steel, for under controlled procedures that minimise the risk of
the sake of least weight, if doing so risks delay and error. It may be a platitude to suggest that the
far more cost. designer should ‘get it right first time’ but it remains
the best aspiration and a company’s QA system
should be geared towards achieving that objective
and certainly towards preventing gross error.
7.2 Managing the design process
Design reviews should be held periodically to check
that the design matches what is required, that it
For any project, there should be a clear ‘basis of ‘looks right’ and to allow peer review of the
design’ that records all the key data and defines how judgements made by the designers.

32 The Institution of Structural Engineers Risk in structural engineering

 Risk in design 7.3

7.3 Clarity of responsibility 7.5 Clarity of design requirements

There is a significant risk on very large projects that Risk reduction starts at concept stage. Questions
the design process becomes so fragmented that no should firstly be asked about the design information:
one party retains a clear overview of the whole. – What exactly is the structure required to do?
Where more than one organisation is involved in – What are the key drivers in design that have to be
design, it is essential to make the division of achieved?
responsibility clear. There should be one engineer – What loads is the structure required to carry?
responsible for overall stability and one lead designer – How much uncertainty is there in these
to set the demands on all subordinate designers. parameters?
Thus, for example, the lead steel designer has to – Is the technology available to match the
define the standard to which connections are aspirations?
designed; the lead concrete designer has to set the – Are the design processes being used proven
standard to which, say, precast units are designed. enough to deliver the goals with confidence?
This is to avoid the risk of misunderstanding which is – Is the form of structure robust and insensitive to the
inherent across the interface. Many clients prefer to accuracy of the design assumptions?
place a concept design contract which terminates – Have all credible modes of failure been considered?
when the design goes to the fabricator or – Is overall stability absolutely dependent on any one
constructor, making direct communication single point?
impossible. The risks in this approach should be – Is the structure buildable within budget?
pointed out to the client; these may include confusion
or inadequate attention to detail or the presumption It is frequently the case that the client or end user
that some preceding party has taken care of does not really know the functional demands or the
important aspects. Equally, there are commercial risks necessary design data with precision, and one of the
to the second party in accepting responsibility for a skills structural engineers should deploy is the ability
concept that might be flawed. If the concept to define the design information required and ask for
designer’s involvement does terminate, overall it. If the information is not available, then mutually
responsibility should be formally handed over after understood bounded information (like worst case
making sure that the recipient is competent to take it loading) should be agreed. All agreements should be
on. Section 5.8 discusses the value of competence recorded in writing.
as a risk management tool.
Most new structures require foundations and there is
In many countries, the law requires certification of the always inherent uncertainty in the assumptions
design. In the USA, design must be supervised by a underpinning the design. As numerous court cases
licensed structural engineer. In England and Wales, will testify, it is a false economy to skimp on site
there is a duty on local authorities. In Scotland, investigations. Every effort should be made to verify
licensed independent engineers may issue essential design assumptions and, if this is not
certificates. The certifying engineer should assess all possible until actual construction, it should be made
aspects and interfaces thereby giving the overview quite clear who is carrying the residual risk. Likewise,
that has sometimes been missing in the past. in refurbishment projects, there is often great
uncertainty about the condition of the existing
There needs to be special care if the programme is structure and, equally often, lack of opportunity to
tight. There should always be adequate programme probe it destructively before alteration on site. If that
time to implement the chosen approach, but there is the case, the design team should have a strategy,
are occasions when rapid action is required to assure supported by sufficient time and money, for onward
safety or to achieve a client’s prime objective. On investigation and for making changes as the picture
those occasions, the design concept should fit the clarifies; there is a risk that gross deterioration will be
time available; it is much less risky to produce simple discovered.
robust designs than it is to attempt the elaborate.

7.6 Designing robust structures

7.4 Design changes
7.6.1 Principles of robustness
Design changes occur on most projects, although
excessive changes during the design development Robustness is a valuable attribute for any structure
phase are a frequent cause of commercial dispute on particularly where the consequences of failure could
cost and delay, posing risk to those who make them be high, for robustness will assist resistance to faults
or those who fail to manage them. There are often and hazards of all kinds. It can be defined as: the
changes in the loading and dimensional information ability to resist unplanned events without
given by other team members to the structural disproportionate or catastrophic failure. Robustness
engineer. To minimise the risk of a drawing passing to is, therefore, useful in desensitising the structure to
construction with the wrong information, there should the consequences of error or unforeseen risk.
be a strict design change procedure. Throughout a Reducing the risk of progressive collapse is one
structure’s life, certainly on major structures, there benefit of robustness and is a design obligation.
should be a design authority with responsibility to
assure the safety of any proposed changes. It is fairly Awareness of progressive collapse developed in the
obvious to engineers that to execute changes safely UK as a result of the Ronan Point collapse in 1968
during the life of the structure, reliable as-built records (see Appendix A.19). More recently, international
are required, but the benefits may need to be recognition of this issue has increased following a
explained to the client who will need to retain them. number of terrorist incidents in which the robustness

The Institution of Structural Engineers Risk in structural engineering 33

7.7 Risk in design

The Institution’s report on robustness7.6 can be

consulted for an elaboration on the principles and
application of structural robustness. This does not
cover specific UK requirements for Class 3 structures,
which require dedicated risk assessments. The
Institution has prepared a further report, Manual for the
systematic risk assessment of high risk structures
against disproportionate collapse7.7. Guidance can be
found in The Structural Engineer7.8.

7.6.2 Designing for accidental loads

Most structures are designed to withstand known

actions such as dead or live load. Some may also
require design for less predictable effects such as
impact or explosion, using a combination of
robustness and energy absorption. Guidance on
accidental load cases and associated design
strategies can be obtained from BS EN 1991-1-
7:2006 Actions on structures. Accidental actions7.9
(EC1 Part 1-7).

Section 3 of EC1 Part 1-77.9 describes design

strategies for identified accidental actions (loads) such
as explosions and impact, and for unspecified causes.
This distinction is very useful. For identified actions, a
hierarchy of design methods is given, aimed at
reducing the probability of occurrence, reducing the
Figure 7.1 Glass plate carrying ’sea’ around ship severity of damage or reducing the consequences of
failure.

For unspecified actions, Annex A of EC1 Part 1-7

of the structure, or otherwise, had a significant effect defines risk classes, also known as consequence
on the level of damage. Stability is often taken to classes, for different types of buildings and levels of
mean resistance to collapse but its full meaning is use. These describe buildings with related activities and
more subtle; the principle being that a minor change occupancy levels (as appropriate) in four broad
in any of the assumed conditions should not cause a groupings which vary from lower risk up to higher risk.
disproportionate change in state. National building regulations may use similar
categories. For each class, rules are defined to limit the
An example is Brunel’s ‘SS Great Britain’ (Figure 7.1), consequences of localised failure.
surrounded in dry dock by an artificial ‘sea’ consisting
of shallow water supported on horizontal glass plates Annex B of EC1 Part 1-7 gives specific guidance on
that span between steel grillage beams7.4. As a brittle risk assessment in the context of accidental loads. In
material, with people walking beneath it, the glass is in the Eurocode, ‘mitigation’ is used to mean risk
a sensitive position, so is much thicker (21mm) than reduction by any means, and not to mean reduction
needed purely on stress grounds. of the consequences if the event occurs, as it is used
in this Report.
It is often possible to provide robustness in a
structure without adding substantially to cost and As a design code, EC1 Part 1-7 gives prescriptive
providing robustness is generally a cheaper and more design rules. It would be very difficult to write such a
effective way to reduce the risk of failure than by design code around the subjective, judgement based
using heavier sections to provide more strength. approach to risk assessment which is recommended
Robustness is not obtained by designing for more in this Report. There is a place for both approaches,
load cases, or by using a higher factor of safety. To depending on the type of risk and the knowledge and
develop robustness when designing either permanent experience of the designer.
or temporary works, consider the following principles:
– Ensure there is capacity for horizontal load
transmittal (including notional loading) and clear
load paths down to ground level. 7.7 Designing for construction
– Use materials and structural forms which are ductile
and whose capacity is insensitive to changes in
load, geometry or material properties. 7.7.1 Buildability
– Tie all the elements of the structure together.
– Provide alternative load paths where possible. A standard structural engineering approach is firstly
– Allow ‘non-structural’ elements to carry load in an to determine a structural form. In making this choice,
emergency. regard should be given to construction needs. Apart
from a basic duty to assure designed structures are
These principles are among those defined in Section ‘buildable’, it may well be that the worst loading
2 of BS EN 19907.5 (Basis of structural design). This is conditions or some particular loading or stability
both a useful summary of the principles of structural condition occurs during site assembly. The risks of
design, and an authoritative reference should one be this being found out too late, or not being
required (for example, to support explanation to a client comprehended by the constructor are that
over the need for robustness). unwarranted safety risks may be imposed (with the

34 The Institution of Structural Engineers Risk in structural engineering

 Risk in design 7.7

liabilities that flow from that) or perhaps the imposition

of extra cost and programme delay which give rise to
dispute. If the construction skills required are at all
unusual, early co-operation with a constructor is a
key risk reduction policy.

Many countries have introduced legal responsibilities

to design for safe construction, and in particular to
co-ordinate design to ensure safety. There is a
perception among some engineers that these duties
can be met by specifying certain routine precautions
(safety nets, reducing component weights etc.). While
these may be valid mitigation measures, they are only
a small part of the story; the real issue is that the
whole design should be properly conceived. While
this Report aims to recommend internationally valid
good practice, not to interpret or restate national
legislation, useful guidance can be found in the
Approved Code of Practice7.10 to the UK’s CDM
Regulations7.11, 7.12. Guidance has also been
produced in the UK by industry and published by
ConstructionSkills7.13; Section 2.6 of which contains
specific advice on designing out risks.

Structures have to be safely ‘buildable’. As a

minimum, the loading and stability states that are Figure 7.2 London Eye
likely to exist during construction need to be
addressed and a strategy devised. There is no firm
rule about how these temporary states should be
catered for, i.e. by strengthening the permanent occurred due to inadequate bearing or lack of
design or by providing temporary works, but a good consideration for overall stability.
design will make a proper judgement and document
the information both to manage commercial risks and A key aspect of ‘buildability’ is the specification of
eliminate or identify site safety risks. These actions achievable tolerances. Disputes and conflicts over
might be as simple as ensuring that the floor capacity lack of realism in choosing tolerances, especially at
allows mobile elevating work platforms (MEWPs) to interfaces, may be perhaps the most common of all.
be used or they might be as sophisticated as In the worst cases these have led to lack of
designing bridge girders for the stress states inherent functionality with repercussions for the design team.
in a cantilever bridge launch. On the one hand, the design team should appreciate
the difficulties of site work and avoid specification of
For anything other than routine structures, there tolerances that are unnecessarily tight or impossible
should be a close connection between the method of to achieve. On the other hand, the constructors
construction and the design of the structure; both in should meet the specified requirements, and discuss
its mode of erection and for the structure’s strength any difficulties with the designer. Where it is
and stability in temporary states. Figure 7.2 shows necessary to define close tolerances, the drawing
the London Eye, where construction was should make clear where and what these are and
fundamental to the concept. Construction sequences include references to any relevant specification
and interim stability states require particular attention clauses.
in construction of deep basements, retaining walls,
steel framed buildings and hybrid structures where, 7.7.2 Designing out construction hazards
for example, the stability of the whole relies on a
concrete core. It is incumbent on all members of the professional
team to be aware of typical site hazards associated
Not all risks are those linked to safety or structural with common construction processes. These include
failure. Commercial design disputes are far more hazards linked to, among other things, excavation,
common. For example, in steelwork design there concreting, steelwork erection and building masonry.
have been many disputes after concept design, The processes of demolition and refurbishment also
where main member sizes have been chosen on the need to be understood.
basis of minimum weight but have proportions such
that connections could not be economically made or There have been incidents of site injury (including
have required excessive stiffening within the death) which could have been avoided by better
connection zone. This risk is exacerbated if, as is design, perhaps as simply as not requiring workers to
typical in the UK, connection design is a separate struggle with the lifting of heavy objects. Some
activity to member sizing with the result that failures have occurred simply because the design
conceptual designers fail to gain adequate team has failed to consider or communicate the risks
experience. In concrete design, there have been properly.
many examples of poor detailing or proportioning
having little regard for the practicalities of placing and Designers should consider every person affected by
compacting concrete around congested the work activity. The key people at risk are the
reinforcement, any of which might lead to loss of operatives rather than their managers, and the best
capacity, excess cracking or poor durability with place to record health and safety information is on
consequent claim. There are cases in precast those working drawings which will be used by both
concrete design where significant failures have construction operatives and maintenance personnel.

The Institution of Structural Engineers Risk in structural engineering 35

7.8 Risk in design

Significant risks can be listed in an eye catching box abstracted for onward dissemination on the
on a related drawing. This is sometimes known as a drawings.
safety, health and environmental (SHE) box. It is also
recommended that information on risk reductions 7.7.4 Designing for unfamiliar environments
taken from the design team meetings and the risk
register together with the drawings describing When designing structures to be built in an unfamiliar
significant risks, are included in the lifetime records environment, such as another country, the design
for the structure. Where the construction information organisation should be competent to do so. This will
is conveyed via computer models using Building require information on local construction methods,
Information Modelling (BIM), specific consideration materials and practices and may require employing
should be given to the way risk information is local engineers. It should not be presumed that
communicated. familiar materials are available. An operation which is
quite safe with a workforce who are familiar with it
Ideally, the design for permanent access to the and well trained may be quite risky if used by those
upper parts of the structure should allow who are unfamiliar. The capabilities of the available
construction to be sequenced so that workers are construction plant may not match what the designer
protected; e.g. so that stairs can be erected in time is accustomed to, with the risk that some operations
to avoid need for temporary access provisions. The might not be feasible; conversely, faced with what is
structure should be designed to support all the for them a novel design, operatives might attempt to
loads which may be placed on it, including use an unsafe approach. Caution needs to be
materials, plant and equipment during construction exercised as the construction quality achievable in the
and installation. See Table 4.1 (Representation of country of design may not be achievable in the
construction loads) of BS EN 1991-1-67.14. country of construction.

The risks to workers are not just injury but also to

health7.15. Designers should avoid specifying materials
that may be harmful to health, and be wary of details
that require construction processes, such as use of 7.8 Design for the whole building life
vibrating machinery. Internationally, there are many cycle
examples of good practice; in Japan, self-compacting
concrete is used extensively, avoiding the need for
mechanical vibration. Reducing risks during the life of a building starts at
the concept and design stages. A key need is to
7.7.3 Communication of risk identify how the structure will age and how materials
may degrade. There can be severe financial and
It is not the purpose of this Report to describe the safety risks to clients and designers consequent on
communication skills that a structural engineer should premature degradation. The financial risks arise from
possess. In terms of risk, however, inadequate data is both the capital cost of repair and from loss of
known to have allowed the wrong materials to be income if the structure becomes non-operational.
used, clashes between concrete and steel, clashes Liability may be high if the causes were foreseeable.
between services and structure, confusion about Designs should consciously address issues of
what was propping what up at any stage, etc. durability and include provisions for safe access for
Production of clear and complete information is just maintenance and for inspection of any areas at
as important as skills in calculation. Advice may be risk7.17. Issues such as corrosion in cable anchorages
found in the IChemE publication, Communication illustrate these concerns and present a good case for
skills for engineers and scientists7.16. giving a higher priority, in the design of certain
structures, to the practicability of maintenance7.18.
The design team should consciously consider the Where degradation is anticipated, for example, by
information they are supplying, with particular fatigue damage, it is most important during design to
attention being paid to data where errors might not identify areas at risk and ensure these can be
be obvious until too late (such as key setting out inspected in service. Design should also address the
data or key materials). They should also ensure safety and health of building users. This might, for
clarity of communication regarding the load paths for example, affect the choice of materials or specification
vertical and horizontal loading, and which elements of cladding materials above areas where many people
provide stability to which; this is especially important will pass. Where the design allows for installation of
in hybrid structures of mixed materials (for example, plant and equipment, it should also allow for safely
where a steel frame might be stabilised by a replacing it during the building life, if this is likely to be
concrete core). necessary.

The identification of hazards and corresponding The designer should ensure that all relevant
evolution of design risk reduction in an overall project information about the facility is available on hand-over
is best achieved via collective team debate. Risk to the owner and/or occupier, including staged
thrives at interfaces. During such debate, reports can hand-over and where there may have been minor
be received from the various designers and decisions works.
agreed for taking the project forward. A failure to
co-ordinate, for example, between architecture,
building services and structure for both finished and
installation phases risks all manner of site and 7.9 Designing for future demolition
operational problems. The results may be a failure of
functionality, fit or even danger if it results in
unauthorised site modifications. Conclusions should Structural designers should consider making
then be recorded in an updated hazard/risk register. recommendations on how best to demolish what
From that register, essential information should be they have designed. The basis of this should be

36 The Institution of Structural Engineers Risk in structural engineering

 Risk in design 7.10

structural engineering philosophy, principles, 7.12 References

background and detail for the design. It should be
clear what provides overall stability to avoid the risk of
that part being taken out prematurely. The benefit of 7.1 ‘Firms fined over screen collapse’. BBC News website,
‘as-built’ information should be available as a legacy 5 June 2008. Available at: http://news.bbc.co.uk/1/hi/
(in the UK, included within the health and safety file), england/west_midlands/7438044.stm [Accessed:
as well as records of alterations made during the 19 February 2013]
structure’s life time. This may include reports on
component degradation. 7.2 Institution of Structural Engineers. Structural design –
the engineer’s role. London: IStructE, 2011

7.3 The Institution of Structural Engineers. Guidelines for

7.10 Procurement and planning the use of computers for engineering calculations.
London: IStructE, 2002

Apart from design itself, other activities during the 7.4 Jofeh, J. and Perry, A. ‘Sea of glass: ‘Refloating’
design phase can have a significant influence on risk. Brunel’s SS Great Britain’. Arup Journal, 3, 2005.
These are not always the sole responsibility of the Available at: http://www.arup.com/_assets/_download/
structural engineer, but it is important to influence download468.pdf [Accessed: 19 February 2013]
them and to ensure that the consequences of the
decisions are understood by those making them. 7.5 BS EN 1990:2002+A1:2005: Eurocode – Basis of
structural design. London: BSI, 2010 [Incorporating
Selection of the right contract form and a competent corrigenda December 2008 and April 2010]
constructor can have a significant influence on risk.
All quality, safety and competence requirements in 7.6 Institution of Structural Engineers. Practical guide to
the main contract should cascade into all levels of structural robustness and disproportionate collapse in
subcontracts. It is often preferable to limit the buildings. London: IStructE, 2010
number of levels of subcontract to avoid too many
interfaces. 7.7 Institution of Structural Engineers. Manual for the
systematic risk assessment of high risk structures
It is preferable for the design contract to remain in against disproportionate collapse. London: IStructE,
force into the construction phase and for construction 2013
contracts to be structured such that appropriate
periodic checking can be carried out to ensure that 7.8 Harding, G. and Carpenter, J. ‘Disproportionate
design intent is being translated into reality. collapse of ‘Class 3’ buildings: the use of risk
Subcontractors carrying out design should be assessment’. The Structural Engineer, 87(15-16),
required to formalise their own basis of design and 4 August 2009, pp29-34
construction method statements and submit them for
review. This does not necessarily transfer liability, but 7.9 BS EN 1991-1-7:2006: Eurocode 1: Actions on
it increases the chances of exposing error in the structures – Part 1-7: General actions – Accidental
constructor’s assumptions and reduces the risk of actions. London: BSI, 2010 [incorporating
gross misinterpretations. corrigendum February 2010]

While the detailed construction programme may be a 7.10 Health & Safety Executive. Managing health and safety
matter for the constructor, the programme at the in construction: Construction (Design and
design stage should aim to sequence activities to Management) Regulations 2007 Approved Code of
eliminate instability and minimise hazards to the Practice. L144. Sudbury: HSE Books, 2007. Available
workers and public. Thus the programme cannot be at: http://www.hse.gov.uk/pubns/priced/l144.pdf
formulated until the construction process and risks [Accessed: 19 February 2013]
are understood.
7.11 The Construction (Design and Management)
Regulations 1994 (SI 1994/3140). Available at:
http://www.legislation.gov.uk/uksi/1994/3140/contents/
7.11 Conclusions and recommendations made [Accessed: 19 February 2013]

7.12 The Construction (Design and Management)

The design phase, particularly the concept design, Regulations 2007 (SI 2007/320). Available at:
offers major opportunities for risk reduction. To use http://www.opsi.gov.uk/si/si2007/20070320.htm
these opportunities, designers should be aware of [Accessed: 19 February 2013]
the uncertainties in design and the practicalities of
construction. Communication and co-operation are 7.13 CITB – Guidance to the CDM Regulations [separate
vital for the reduction of risks to health and safety guidance documents for Clients, CDM Co-ordinators,
over the life cycle of the structure, as well as Designers, Principal Contractors, Contractors and
disruption to the project as a result of bad design. Workers]. Available at: http://www.citb.co.uk/en-
Where designers have a specific legal responsibility to GB/Health-Safety-and-other-topics/Health-Safety/
design out risk, as under the UK’s CDM health-safety-legislation/ [Accessed: 12 March
Regulations7.12, this includes risk from structural 2013]
design errors just as much as risk from failing to give
thought to safe construction. The detailed advice in this 7.14 BS EN 1991-1-6:2005: Eurocode 1: Actions on
chapter should be read and understood by all structures. Part 1-6: General actions – actions during
engineers who carry out design of any kind, whether as execution. London: BSI, 2010 [Incorporating
part of the main design phase or not. corrigendum July 2008]

The Institution of Structural Engineers Risk in structural engineering 37

7.12 Risk in design

7.15 Institution of Civil Engineers. Health and safety

guidance and legislation. Available at: http://www.ice.
org.uk/topics/healthandsafety/Guidance-and-legislation
[Accessed: 19 February 2013]

7.16 Institution of Chemical Engineers. Communication skills

for engineers and scientists. 4th ed. Rugby: IChemE,
2007

7.17 Iddon, J. and Carpenter, J. Safe access for

maintenance and repair: guidance for designers.
C686. 2nd ed. London: CIRIA, 2009

7.18 Health and Safety Executive HM Railway Inspectorate,

An Assessment by HSE of the structural integrity of the
Forth Rail Bridge report, C10. Sudbury: HSE Books,
1996. Available at: http://www.railwaysarchive.co.uk/
docsummary.php?docID=2107 [Accessed: 19 February
2013]

38 The Institution of Structural Engineers Risk in structural engineering

8 Risk management during construction

8.1 Introduction construction, while in most cases it should receive

more.

This chapter describes actions that should be taken

during construction to manage risk. The most serious
risk is collapse of the structure or temporary works, 8.3 Resourcing and planning
or harm to construction workers. On many projects,
however, the most frequent risk is that errors in
structural engineering or its communication lead to Management of risks during construction starts, as
cost or programme over run. Action taken during the at any time, with hazard identification. The
design phase to minimise risk in construction is construction work should be planned in terms of
covered in Chapter 7. both how and when each activity will be executed,
making sure it is then resourced in terms of
competent labour, time and finance. The number of
people employed, and their skills, should permit
8.2 Causes of incidents adequate planning, checking and supervision. If the
project is rushed, or under-resourced, short cuts will
be taken and both quality and safety will suffer. This
Construction is a complex activity, requiring the does not mean that projects cannot be built to a
integration of many activities across a wide range tight budget and timescale, just that there should be
of disciplines and the translation of an abstract a realistic plan in place setting out the required
‘design’ into reality. The underlying presumption at objectives and how they can be achieved. That plan
the start of any site work should be that it will be should be framed around a conscious assessment
risky: things can go wrong. Most experienced of the hazards, including those outside the
engineers know that projects rarely go entirely to constructor’s control such as adverse weather or
plan. The causes (in effect, the hazards) may be in unforeseen ground conditions.
the design or its communication, such as errors on
drawings, misinterpretation of specifications, Ensuring that materials are delivered to site in the
design changes, failure to eliminate hazards etc., all right sequence and quantity will not only save money
of which raise the prospect of rework, wasteful but will reduce risks to both health and safety of
expenditure, faulty construction and danger to life those involved and project completion. At the delivery
or health. The hazards may be in errors by the site points there need to be proper facilities for offloading
staff and suppliers (e.g. incorrect setting out, the and storing. Surplus material lying around a site can
use of a wrong concrete mix, missing be one of the main contributors to accidents, either
reinforcement or just bad workmanship). Some of as an obstruction, a toppling stack or, if stored on the
the greatest risks on site arise from unplanned structure, an overload. The risk of deviating from the
activities. planned erection sequence is reduced if deliveries to
site follow the ‘just in time’ principle, with a suitable
Examples of the risks which can be directly buffer to avoid delay.
minimised by good planning and management on site
include excessive construction loading on floors not The designers of the permanent works should have
designed for it, improper sequencing of construction, ensured that there is at least one safe way that the
material weakness and temporary structural structure could be erected and that the overall
instability. Most structures are more vulnerable to stability systems are clear. If the structure is
instability when in the assembly stage than when unconventional, or the design anticipates a specific
complete. Temporary works, e.g. falsework and erection method, a description should have been
excavations, often receive less attention than provided. If drawings are not clear, the designer
permanent works, even though the loading and should be asked. If the design contract has been
construction process may be more onerous but less terminated this may require specific arrangements by
certain. The immediate safety of workers on site can the client. This does not necessarily mean that the
be at risk from working at height or in confined structure has to be erected the way envisaged by the
spaces and handling or being too close to heavy designer, for the constructor will probably have more
objects and machinery. Their short or long term experience in this area and is entrusted with
health can also be at risk from manual handling, construction for that reason.
vibration from tools, asphyxiation, infection and
hazardous substances. These include not only Planning construction should be based on the broad
materials like asbestos which are still found in existing strategy of making sure the vertical and horizontal
structures, although banned in new construction in loads can be carried at all times and that stability is
most of the world, but also common materials, such assured at all times. Thereafter it should be assured
as concrete, that cannot easily be eliminated8.1. Hot that all parts are safe to lift and stay stable while
work such as welding8.2, flame cutting or melting being lifted and that there is safe access for workers
bitumen brings with it obvious personal danger but also to locations inherent in the assembly sequence. After
the risk of fire. Some of the biggest fires on record the broad strategy is determined, a detailed appraisal
(Broadgate8.3, 1990) have occurred not in finished of each activity needs to be made identifying any
buildings but during the construction phase. relevant risks and making sure the workers are
Refurbishment often receives less attention than new properly briefed. Documented method statements,

The Institution of Structural Engineers Risk in structural engineering 39

8.4 Risk management during construction

safe systems of work and relevant QA procedures proprietary suppliers, etc. On major projects, a third
should be produced, and checked. party role to assess construction methodology can
be a powerful insurance policy.

Adequate welfare facilities give staff and operatives the

opportunity to maintain hygiene standards; apart from
8.4 Competence, management and the direct benefits, this helps create a safety culture.
welfare

A key difference between a construction site and 8.5 Communication

almost any other workplace is that a site is always
changing. A factory can be set up with safe working
arrangements for operation and maintenance and Good communication is vital, including both
employees can learn to use these. On construction communication from the design office to the site (for
sites it is necessary to review the safety procedures both permanent and temporary works) and within the
and protective measures continually as construction site itself. Such communication can be by traditional
progresses. Not only does each site vary over time, drawings and documentation, or using electronic
but different sites require a mobile workforce. As a Building Information Management, but direct dialogue
result, the construction workforce often has a high (at all levels) is strongly recommended. Vast amounts
turnover and includes at least some who have been of ‘written communication’ carry the risk that vital
unable to find other employment. data will be lost in information overload. Constructors
and designers should work together, preferably with
Maintaining competence in such a dynamic situation the client as well, to ensure that the right information
requires continuous monitoring and proactive work to is produced and conveyed, thus making the site work
ensure that all workers have appropriate training and, efficient, rapid and safe. Any changes to the work
if not experienced, supervision. Every person working shown on the drawing should be referred back to the
on site should have both general safety awareness designer, since the full reasons behind the original
training and additional training proportionate to their design may not be appreciated on site.
site responsibilities. As a minimum, everyone in the
industry should be aware of common hazards and Communication between parties on the site itself is
such risks as the dangers of confined spaces and of crucial. Accidents have occurred due to workers
visiting sites alone. Those who are directly involved misunderstanding drawings and carrying out work
with structures, such as steel erectors, scaffolders without checking with a foreman. Construction labour
and supervisors should also understand the has always been mobile. Where a proportion of the
principles of stability. All workers should be workforce are not native speakers of the language in
encouraged, or required, to hold an appropriate use, this presents its own obvious risks. Lapses in
safety qualification, if one exists. In the UK, the communication between different contractors and
Construction Skills Certification Scheme8.4 issues subcontractors on the same site have also led to
cards to certify that the holder has demonstrated a accidents, often due to failure to notify others of
knowledge of safety appropriate to their role on the site. hazardous activities. The role of principal contractor in
Staff visiting sites (especially if intermittently) should be the UK’s CDM Regulations8.5 was partly intended to
given inductions on site specific risks that can change combat this issue.
on a daily basis, such as evacuation routes.

Construction projects range from minor building

alterations to major bridges. Each brings with it a 8.6 Loading conditions
particular set of hazards and is carried out by a team
whose skill levels and experience clearly differ. Large
projects will be supported by a full range of specialist Critical loading conditions may exist during the
roles and skills. On smaller projects, these roles may construction phase that are entirely linked to the
need to be combined, or specialists may need to be manner of construction. For example, wind loads on
brought in temporarily. To minimise the risks, a bare and incomplete steel frames or large panels of
combined effort is required by the designers and reinforcement and shuttering, construction loads that
constructors to make sure the pool of experience exceed the permanent loads and wheel loading from
assigned to the project has the best chance of vehicle or crane access are all common issues. It is
spotting what might go wrong. During construction, important when planning construction to establish
adequate supervision by appropriately skilled staff is whether the design intent for the permanent
required and that might include ongoing advice from members allowed for the particular erection method
the design team. proposed. In small scale work, such as
refurbishment, this may be just to ensure that existing
A proper management structure will support a team members have not degraded. In general, the task is
approach as the safest approach for avoidance of all to make sure that the evolving structure is capable of
manner of risks. Within the team it should be quite taking the loads at each stage. This may include
clear who is responsible for what, at each stage. The ongoing assessment of the actual rather than the
main contractor’s role and responsibility for overall expected condition or loading. Major project events
co-ordination of site safety is absolutely fundamental. may involve a single sequence of heavy lifts, such as
This role is particularly vital when there are numerous raising the London Eye8.6, the Wembley Stadium
subcontractors; the main contractor has to control Arch8.7, 8.8 or launching a bridge. Depending on the
the interfaces to meet the twin objectives of making method chosen by the constructor, such events can
sure what is built satisfies design intent and that stress the structure into states not fully appreciated by
safety is assured. This includes liaison with temporary the designers for the permanent condition. One
and permanent works designers, specialist example of a failure during construction occurred to the

40 The Institution of Structural Engineers Risk in structural engineering

 Risk management during construction 8.7

tunnel construction at Gerrards Cross8.9 in 2005 (see

Figure 8.1).

The tunnel was to allow backfilling over an existing

railway to provide the site for a supermarket. The
loading conditions applied to the precast arch units
would have differed during progressive backfilling
from those which would exist after backfill had been
completed, although this may not have been the
cause of the collapse. There were no injuries, but the
client had to pay substantial compensation to the
railway operator for obstruction to the railway and
was not covered under its public liability
insurance8.10.

Permanent works designers often have great difficulty

envisaging the loading conditions that may exist
during construction and so the construction engineer
on site should always verify their proposed
methodology. Construction loads may arise, for
example, from personnel and hand tools, storage of
construction materials, non-permanent equipment,
moveable heavy machinery and equipment, Figure 8.1 Tunnel at Gerrards Cross
accumulation of waste materials, or loads from parts
of a structure in a temporary state. These may exist stiffness that concrete will eventually provide. In all
when material such as concrete is in an immature construction, it is normal for the main stabilising
state. A typical case exists when upper concrete element (such as a core or braced bay) to be
floors of a building are cast supported off the floor constructed first and then for the rest of the structure
below (perhaps propped through multiple floors). to be progressively erected away from that point. For
various reasons, however, the chain of structural
integrity back to stability points might be interrupted
during construction.
8.7 Sequence of construction
In complex projects, a dedicated monitoring regime
can be established to assure that all is going to plan.
The preparation of drawings or 3D models showing a In full size structures it is not practical to measure
structure in its incomplete stage(s) should be stresses directly. It is possible, however, to survey
encouraged, particularly for innovative, complex or displacements and map those against predictions for
unusual structures. These should be studied to each stage. When there is correlation, it is an
understand where the structure obtains strength and indication that the stresses are acceptable. Such
stability. For example, a steel truss may be laterally monitoring is essential with certain procedures,
stabilised by purlins in its final condition but can only notably the excavation of deep basements or
be erected with those purlins absent. Bridge beams tunnelling. Indeed it is central to the success of NATM
stabilised by a concrete deck will have to support the (New Austrian Tunnelling Method) techniques8.11. The
weight of wet concrete while lacking the in-plane illustrated failure at Heathrow8.12, 8.13 (see Figure 8.2)

Figure 8.2 Recovery from tunnel collapse, Heathrow, UK

The Institution of Structural Engineers Risk in structural engineering 41

8.8 Risk management during construction

resulted at least in part from lack of management of the better to eliminate the hazard or use collective
ground survey information that was collected. methods of protection that individuals cannot decline.
It is human nature to try to avoid things that make the
An increase in prefabrication means that building parts job more difficult, particularly if the worker has limited
can arrive on site fully formed and only require putting understanding of the risk. PPE is often considered to
into place. The structure may not, however, be stable be an encumbrance so the need for it should be
in some intermediate forms, requiring care in the way avoided whenever possible.
individual parts are lifted and temporarily supported.
An erection sequence should be devised accordingly. There are, however, situations where PPE can
provide a real risk reduction, such as asbestos
As well as ensuring stability, selection of the removal and other instances of contamination. In
construction sequence provides an opportunity to these cases, the hazard is clear and workers will
improve construction access, for example, by insist on wearing it. The focus in such cases is
constructing stairs and handrails early or by ensuring that it is adequate.
pre-assembling complex items at ground level.

8.10 Access and work areas including

8.8 Temporary works working at height

Many temporary works failures have been recorded, There are clear risks if working areas are not provided
such as Nicoll Highway8.14, 8.15 (2004). The design of with safe access and egress. Falls from height are still
temporary works can be more complicated and risky numerically one of the greatest sources of injury.
than design for permanent works. The design loads Working platforms should normally be provided, but
and the conditions of installation are often much less there are no universally right or wrong answers.
certain and the measures for the provision of stability Scaffolding is the traditional way to provide platforms,
can become confused. Temporary works should be but its erection involves a risk to the scaffolders. For a
designed properly, not simply erected based on one-off task that does not require carrying of tools or
someone’s experience8.16. Similarly, foundations for materials, a carefully managed ladder may offer least
temporary works should not be ignored or treated overall risk. For short duration work in difficult areas,
lightly. The appointment of a temporary works roped access may be the safest approach, but this is
co-ordinator, as defined in BS 59758.17, should be true only if proper training and equipment are
considered as a means to address any temporary/ provided.
permanent works interface issues.
As with temporary works, there are many examples
Falsework, in particular, has been the cause of many of access scaffolding collapsing or being unsafe to
incidents. Awareness of the dangers seems to be work from. Tools and equipment need to be
cyclical. Following a number of accidents in the UK, the maintained and readily be available so that a ‘make
Bragg Report8.18 in 1974 was the trigger for the do’ attitude is avoided. The configuration of simple
production of BS 59758.17, which set new standards for access scaffolds is within the skill of a trained and
falsework, but concern continued. BS 5975 was experienced scaffolder, but complex, free standing or
updated in 1996 and an HSE construction information sheeted scaffolds require design by a structural
circular (CIS568.19) was issued to improve awareness of engineer. Collapses have occurred where access
the issues. A SCOSS Topic paper8.20 notes that scaffolds have been used as temporary works, such
increased use of proprietary systems and subcontracting as to restrain a façade.
has reduced the knowledge base among main
contractors and that communication between designers The correct use of a mobile elevating work platform
and erectors can be poor. As part of European standards (MEWP), sometimes called a ‘scissor lift’ or
normalisation, BS EN 128128.21 was issued in 2004 and ‘cherrypicker’, provides a safe working area without
revised in 2008. The UK National Foreword to the need for scaffolding, but requires planning of the
BS EN 12812 points out that it does not include two construction sequence. A MEWP may be unsuitable if
recommendations of the Bragg Report, included in the site is crossed by trenches, the floor slab is being
BS 5975, namely a minimum lateral stability force and cast at the same time, the floor capacity is
appointment of a temporary works co-ordinator, and inadequate or the MEWP would be at risk from
recommends use of BS 5975 for Class A falsework. This vehicle impact.
is an area where published material overlaps and up to
date information should be sought. Similarly, the need to work in confined spaces can be
minimised by careful planning. For example,
installation of mechanical equipment into a tank is
often carried out from the top, which is inefficient and
8.9 Protective equipment requires complex arrangements to protect against
asphyxiation and allow escape. Leaving out part of a
wall until the plant is installed may save time and
Personal protective equipment (PPE) has a high money, as well as being safer.
profile as a risk reduction measure, which is probably
inconsistent with its real importance in the risk
reduction hierarchy. Many countries, including the UK,
place a legal obligation on employers to provide it 8.11 Lifting
and certain items are usually made mandatory for
workers. Nevertheless, partly because of the difficulty
in persuading people to use it, protective equipment Lifting anything is always risky, whether by human or
should be thought of as the last resort. It is much mechanical means. Manual lifting poses a risk of

42 The Institution of Structural Engineers Risk in structural engineering

 Risk management during construction 8.12

operatives sustaining back injury, with an added risk 8.13 References

of trapped fingers or feet. It is essential to have
appropriate equipment and training and to limit the
load lifted. Crane overturning is a fairly frequent 8.1 Institution of Civil Engineers. Health and safety
occurrence and is typically caused by overload or due guidance and legislation. Available at: http://www.ice.
to location on bad ground. All manner of minor lifting org.uk/topics/healthandsafety/Guidance-and-legislation
devices can be misused. While it is preferable to use [Accessed: 19 February 2013]
mechanical aids than rely on human effort alone it is
even better if lifting can be avoided. For example, 8.2 Puybaraud, M-C. and Barham, R. Addressing the risk
concrete can be pumped rather than skipped. of fire during the Construction/refurbishment process
by better management. Paper presented at COBRA
Lifting cannot be avoided completely. Prefabrication ’97, Portsmouth, 10-12 September 1997
might require large lifts, albeit fewer of them but large
lifts are likely to be treated more seriously. All major 8.3 Fire Safety Engineering Consultants Ltd. Structural fire
lifts should be planned in detail and that includes engineering: investigation of Broadgate Phase 8 fire.
assuring that cranes are used within their capacity SCI Publication 113. Ascot: SCI, 1991
(for both load and wind speed) and are supported on
stable ground. This includes proper design of 8.4 Construction Skills Certification Scheme (no date)
attachment points to the parts being lifted and Available at: http://www.cscs.uk.com/ [Accessed:
controlling and managing the lift under supervision. 22 July 2013]

The potential for instability and the need to support 8.5 The Construction (Design and Management)
large and heavy objects from cranes combine to Regulations 2007 (SI 2007/320). Available at:
make erection one of the biggest causes of http://www.opsi.gov.uk/si/si2007/20070320.htm
accidents. Advice can be found in a SCOSS topic [Accessed: 19 February 2013]
paper8.22.
8.6 Berenbak, J. et al. ‘The British Airways London Eye.
Part 2: Structure’. The Structural Engineer, 79(2),
16 January 2001, pp19-28
8.12 Conclusions and recommendations
8.7 Mann, A.P. ‘Design and Fabrication of the new
Wembley Stadium Roof.’ Presentation to the
– Take account of the generic guidance in Chapter 5. Manchester Association of Engineers, 16th November
2006. Available at: http://www.mae.uk.com/
– Establish procedures for managing, monitoring and Wembley%20Stadium%20with%20Pictures.PDF
updating risks throughout the project. [Accessed: 5 March 2013]

– Identify the safest reasonable approach to each risk 8.8 Bizley, G. ‘In detail: Wembley Stadium arch’. Building
for the site concerned – do not assume that what Design, 10 June 2005, pp24-27
worked on one site will work on another.
8.9 ‘Backfill operation probed in Gerrards Cross tunnel
– Allocate responsibility for managing risk to the party collapse’. New Civil Engineer, 7 July, 2005 [As at Dec
best able to do so, and ensure that this is 2012, no formal investigation is believed to have been
understood. completed]

– Plan the work, based on the information provided 8.10 Tesco Stores Limited v Constable and Others
by the designers and with significant input from the [2008] EWCA Civ 362. Available at: http://www.
constructor’s skills and experience, to ensure judgmental.org.uk/judgments/EWCA-Civ/2008/
stability and safe access. %5B2008%5D_EWCA_Civ_362.html [Accessed:
20 February 2013]
– Seek the opinion of the designers in review of the
proposed approach. It is recommended that design 8.11 Nicholson, D. et al. The Observational method in
contracts allow for this. ground engineering: principles and applications. CIRIA
Report 185. London: CIRIA, 1999
– Produce method statements and safe systems of
work and require subcontractors to produce them 8.12 Health and Safety Executive. The Collapse of NATM
for their activities. Tunnels at Heathrow Airport, 20/21 October 1994.
Sudbury: HSE Books, 2000
– Use appropriately competent people to assess the
stability of temporary works, including falsework 8.13 Standing Committee on Structural Safety. The Collapse
and access scaffolds. of NATM Tunnels at Heathrow Airport. SCOSS Failure
Data Sheet SC/06/101. Available at: http://www.
– In major contracts, or where construction quality is structural-safety.org/topicpapers [Accessed:
critical, consider third party accreditation for the 20 February 2013]
construction methodology (and risk management)
arrangements. 8.14 Magnus, R. et al. Report on the incident at the MRT
circle line worksite that led to the collapse of the Nicoll
All this should be aimed at ensuring that the risk of Highway on 20 April 2004. Singapore: Ministry of
gross error is minimised, that the risk of instability and Manpower, 2005
overloading at any stage is minimised and that no
worker or subsequent user of the structure is 8.15 Standing Committee on Structural Safety. The Collapse
exposed to unnecessary risks to their safety or of the Nicoll Highway on 20 April 2004. SCOSS
health. Failure Data Sheet SC/06/102. Available at:

The Institution of Structural Engineers Risk in structural engineering 43

8.14 Risk management during construction

http://www.structural-safety.org/topicpapers
[Accessed: 20 February 2013]

8.16 Carino, N.J. et al. Investigation of Construction Failure

of the Riley Road Interchange Ramp, East Chicago,
Indiana. NBSIR 82-2593. Washington, DC: National
Bureau of Standards, 1982. Available at: http://www.
nist.gov/customcf/get_pdf.cfm?pub_id=908826
[Accessed: 20 February 2013]

8.17 BS 5975:2008+A1:2011: Code of practice for

temporary works procedures and the permissible
stress design of falsework. London: BSI, 2011

8.18 Department of Employment and Department of the

Environment. Falsework: interim report of the Advisory
Committee on Falsework. London: HMSO, 1974 [Final
report, known as the Bragg Report, published as
Health and Safety Executive. Final report of the
Advisory Committee on Falsework. London: HMSO,
1976]

8.19 Health & Safety Executive. Safe erection, use and

dismantling of falsework, HSE Construction Information
Sheet 56. [s.l.]: HSE, 2003. Available at: http://www.
hse.gov.uk/pubns/cis56.pdf [Accessed: 20 February
2013]

8.20 Standing Committee on Structural Safety. Falsework:

Full Circle? SCOSS Topic Paper SC/T/02/01. Available
at: http://www.structural-safety.org/topicpapers
[Accessed: 20 February 2013]

8.21 BS EN 12812:2008: Falsework – Performance

requirements and general design. London: BSI, 2011

8.22 Standing Committee on Structural Safety. Erection of

steel structures: learning from experience SCOSS
Topic Paper SC/T/04/01. Available at: http://www.
structural-safety.org/view-report/scoss174 [Accessed:
20 February 2013]

8.14 Bibliography

Health and Safety Executive. Benefits and costs. Available at:

http://www.hse.gov.uk/costs/costs_of_injury/costs_of_injury.asp
[Accessed: 19 February 2013]

Mann, A.P. ‘Construction safety: an agenda for the profession’.

The Structural Engineer, 84(15), 1 August 2006, pp28-34

44 The Institution of Structural Engineers Risk in structural engineering

9 Risk management during the life of a structure

9.1 Introduction 9.3 Risks during the life of a structure

This chapter describes considerations for risk Structures may require the attention of a structural
management during the life of a structure, including engineer during their life for a number of reasons.
alterations, maintenance and degradation Principally, these are proposed changes of use,
mechanisms during life9.1, 9.2. There is a strong link modifications or when concern arises over the
with asset management9.3 and an appropriate structural condition. Many facilities will undergo
reference9.4 is given for further reading. Also the effects modification, typically at intervals of about five years.
of construction activities9.5 on existing structures This may be due to deterioration or accidental
should be explored9.6. damage, including fire. Modifications may be required
to meet statutory obligations, such as access for
disabled people or changes to fire regulations. Every
modification involves risk, not just from the
9.2 The life cycle of a structure practicalities of working safely but for the potential for
inadvertent changes to the design intent. When
engaged to review an existing structure, the engineer
Structures have a longer life than almost all other should be alert for any ad hoc or informal changes to
human artefacts. During their life, society changes the structure or the loading, as well as those which
around them and they are often modified for new have been documented. Building occupiers may also
uses. Humanity cannot afford, in terms of seek advice on maintenance; either to maintain or
sustainability, to build new structures where existing extend the life of the building, to address perceived
ones can be used. New uses, however, and new problems, or as a matter of good practice, possibly
loadings from existing uses, will introduce new risks, to meet their obligations as landlord or tenant.
both during the process of modification and when in Whatever the initial involvement in the structure,
use. Structures will also require cleaning and other however, the engineer should consider making
maintenance. Good design will make provision for proposals for its future management. A client who
this, but circumstances change and the provisions has already had to call in an engineer is likely to be
made when the structure was new may no longer be receptive to such advice.
appropriate.
Ensuring structural adequacy is the most important
Significant numbers of bridges on UK motorways part of risk management during the life of a structure.
have had to be upgraded because their This may entail adhering to regulatory9.7
vulnerability to vehicle impact has proven to be too requirements, either for buildings in general or for
great. This was not a consequence of errors in the specific uses. Taking the UK as an example, Building
design, or of experience leading to changes in Regulations are not generally retrospective with regard
design practice, but of society changing its view of to structure, but if there is a change of use or a
which risks are acceptable. Risk assessment in significant modification, they will apply to an existing
these situations is particularly difficult, as the building. In particular, the requirements to resist
cost/benefit ratio for reducing the risk by modifying disproportionate collapse may apply; see Approved
the structure is quite different to what it would have Document A9.8 to the England and Wales Building
been at the time of the original design, yet there is Regulations and Part C9.9 of the Scottish regulations.
pressure from society to meet the ‘modern See Section 7.6 for a discussion of robustness and the
standard’. provisions of BS EN 1991-1-79.10. There are, however,
many ways to improve the robustness of an existing
Managing structures and facilities effectively and structure, particularly if it is required (by the client) to
safely during their lifetime can provide economic meet modern standards or if there are legal
benefits to the owners and users, at many levels, not requirements applicable in the location concerned.
least in mitigating the cost of the dramatic losses
(that society pays for via insurance) if incipient Every part of a structure will decay from the moment
degradation is not detected and acted upon early it is completed. Both the construction industry and
enough. This is recognised in many business society in general have had to face up to the risks of
communities and a ‘new’ profession of facilities concrete9.11 decay (of various forms) and to absorb
management evolved towards the end of the 20th substantial maintenance costs as a result. In some
century. This was partly because owners and cases, as in the Montreal bridge collapse9.12, 9.13
occupiers are realising that management of risks (2006) in Canada and the Stewarton railway bridge
associated with the building or structure can be collapse9.14 (2009) in Scotland, unobserved decay has
addressed as part of overall risk assessment for led to collapse. This may be of the whole structure such
business continuity. Provided the facilities manager as of Pipers Row Car Park9.15 (1997) or partial as in the
has the competence to manage structures, this is a failure of cavity walls from tie corrosion. Each year,
‘win-win’ strategy, addressing a potential liability to people are injured (and sometimes killed) simply by
staff and the public at the same time as protecting parts falling off buildings9.16, 9.17, 9.18. Monitoring can
the business. In most cases, the cost of a building’s avoid unexpected failures but even so the economic
contents or the cost of disruption linked to loss of costs can be high, sometimes extremely so, as
function is far more than the cost of the structure evidenced by the consequences of hanger and main
itself. cable deterioration on the Forth9.19 and Severn9.20, 9.21

The Institution of Structural Engineers Risk in structural engineering 45

9.4 Risk management during the life of a structure

suspension bridges. Disasters such as the King Cross – Fire resistance and escape.
underground fire9.22 (1987) reveal the vulnerability of – Deterioration, corrosion and ageing.
existing infrastructure. The safety lessons from all such – Dynamic and/or fatigue effects.
events9.23 need to be spelt out9.24, 9.25 and applied to – Localised failure/collapse.
other facilities if further events are to be avoided. – Progressive/disproportionate collapse.
– Serviceability failures, subsidence, falling debris.
Even recent history tells us that understandings – Loss of use.
change. Thus Ronan Point9.26 (1968) alerted – Legal liability.
engineers to the danger in certain building types of lack
of robustness and possible disproportionate collapse. – What has changed?
The collapse was initiated by the ignition of gas with a – Is usage and loading as designed for, including
resulting overpressure, a hazard which had not been external factors like climate change9.29?
appreciated in design. Following the failure of – Have standards changed?
structures built using high alumina cement (HAC) – Has deterioration occurred as a result of ageing
concrete in the 1970s, the industry has had to assess or misuse?
significant stock in the light of better understanding – Have new hazards arisen?
about HAC degradation. – Is there new learning from similar structures?

Provisions made by designers are not always – What level of risk is acceptable?
understood by later occupiers, and may require – Level of degradation in the fabric?
reinstatement. For example, it has been known for – Is proactive or reactive maintenance required?
movement joints to be filled with solid material during – How much can the client’s business afford (short
decoration, and ‘strengthening’ to be added (on the term survival may outweigh long term benefit)?
direct instructions of the owners) to prevent the – What are the insurance implications?
inevitable cracks.
– What measures are needed to ensure this?
– Resources to be committed to implementing the
policy?
9.4 Risk management strategy – Time scale for action?
– Who should be involved?
– Interval before next review?
Management of risks during the life cycle of a building
should have been considered at the concept and
design stages (see Section 7.8). The initial strategy
set out by the designers, however, will often need to 9.5 Maintenance regime
be modified, for the same reasons that the building
itself may need alteration9.27, i.e. to make it safer and
more resource efficient to operate9.28 and to take An essential part of any risk management strategy will
account of changing circumstances. be to put in place a maintenance system. It should
define what maintenance and inspection is required,
Risk management strategies will be required for how and when it should be done and which aspects
activities such as refurbishment (long term), should be given particular attention. Although this
maintenance such as painting (medium term) and should be realistic, to ensure that planned activities
activities such as window cleaning (short term/cyclical). are actually carried out, maintenance and inspection
are not optional. If clients are short of funds,
To relate building risks to business risks, the strategic maintenance is often seen as an area which can be
position of the building within the organisation should squeezed with no immediate consequences. Lack of
be considered. The intended uses and occupancy maintenance, however, merely prolongs and
patterns of the building should be considered in the exacerbates a problem which is bound to exist, with
context of the management arrangements. This will increased long term costs. Engineers involved in
enable the appropriate responsibilities to be put in maintenance work should not accept work which is
place for the various aspects of ensuring continuing inadequately resourced to the point where it cannot
safe and efficient use, both in and around the be done safely.
building. Risk management strategies should be
implemented through project and operational risk Failure to carry out adequate maintenance has been
assessments that will take the strategic position into implicated in several major collapses, such as the I-
account. Clients may tend to view any activity in 35W bridge in Minneapolis, USA9.30 and the Malahide
terms of short term cost/benefit, without fully viaduct in Ireland9.31. A contributory factor in the I-35W
understanding the long term risks. collapse was lack of redundancy in the original design.
In other cases, the risk of lack of maintenance may be
The following points should be considered when economic loss (see Figure 9.1).
establishing a risk/hazard management strategy:
– Who may be affected? Different components of a structure have different
– To what should the policy minimise risk? (e.g. to characteristics and will require a different inspection,
human life, to the structure, to business approach and maintenance cycle, for example:
continuity) – Structural materials: concrete, steel, aluminium,
– To whom should the policy minimise risk? (e.g. timber, glass, masonry, etc.
people in/on the facility, people outside the facility, – Façade: cladding, curtain wall, external wall tile and
the owners or operators) renders, windows, structural sealants, etc.
– Bearings, fire stops, movement joints, glass
– What are the areas of risk? balustrades etc.
– Stability in normal use and in extreme events – Non-structural elements: ledges, air-conditioner
such as storm or flood. hoods, equipment/plant supports, etc.

46 The Institution of Structural Engineers Risk in structural engineering

 Risk management during the life of a structure 9.6

It should be possible to inspect, throughout its

lifetime, any metal structure subject to varying loads,
to assess the likelihood of fatigue damage, unless it is
clear from the design that, even if abused, the
number of cycles would not lead to damage during
any credible lifetime.

Provisions made for maintenance may become out-

of-date. Roof access is a good example; originally,
this was the preserve of steeplejacks, who would
climb over slate roofs with no protection. Provision for
maintenance was brought in with the advent of flat
roofs, with parapets or handrails. However, not all
roofs were suitable; some which looked accessible
on first glance were fragile, and people fell through.
Standing seam aluminium roofs were initially provided
with attachment points for harnesses, but these are
no longer a preferred method of protection. Current
thinking9.32 is to design for no maintenance or, for
relatively low-rise structures, to provide a suitable
surface at ground level for access from mobile
platforms. When considering an existing structure, the
maintenance strategy should be reviewed to determine
whether the approach is still optimal, given the
circumstances. In many cases, it will not be practical to
use the method which would be ideal for a new
structure; equally, it may no longer be justified to use Figure 9.1 Collapsed I-35W bridge
the existing methods.

Relevant information may be found in, for example: should ensure that the structure which has been
– maintenance manuals designed is safe. It does not follow that an existing
– statutory records (e.g. in the UK, the health and structure which does not meet modern design codes
safety file), which should be kept up to date. is necessarily unsafe; it may be, or it may not be.

Operating manuals, including health and safety files For example, many structures in the UK which have
should be considered for updating, depending on the been designed to BS 81109.36 may not meet all the
nature of the activities undertaken. requirements of BS EN 19929.37. The change to
Eurocodes was made initially to harmonise European
Maintenance should be planned so that it can be trade rather than for structural reasons, and does not
carried out safely. Much of the guidance in Chapter 8 necessarily mean that structures designed previously
on construction risk may also apply to maintenance. are unsafe. On the other hand, structures designed to
Typical risks to people doing maintenance and also meet the wind loading requirements of CP3,
those who may be affected by their activities include: Chapter V9.38, will quite possibly not meet the
– access (modern bridges often include access requirements of BS 6399, Part 29.39 and
gantries which themselves require maintenance9.33) BS EN 19919.40, and their safety may need to be
– confined spaces assessed. These codes are based on a more
– hot materials (fire risk), toxic materials advanced understanding of wind effects and, for
– falls from height, including fragile roofs. example, their wind and snow maps have been
redrawn.
A serious accident occurred during modifications to
the Avonmouth Bridge9.34 using the access gantry It is for structural engineers to use their skills and
(1999). The gantry was not anchored adequately and judgement to decide whether or not an existing
blew off the end of the supporting beams, falling to the structure is adequate for its proposed use, taking all
ground with the loss of four lives. It is possible that this the relevant factors into account9.35. In the era of
accident might not have occurred if more thought had sustainable development, society cannot afford to write
been put into the way the gantry was operated. Advice off existing buildings just because engineers are not
on the use of such gantries is given in the Institution’s prepared to make judgements in the application of their
report9.33. skills. Judgement, however, is not guesswork, and
should be supported by logic and evidence. This may
be based on pragmatic first principles engineering and
may also make use of the statistical concepts
9.6 Appraisal and assessment described in Chapter 6.

Appraisal of existing structures is a topic on its own;

risks can be minimised by using appropriate 9.7 Conclusions and recommendations
guidance such as the Institution’s guide9.35.

Section 5.6 discusses the use of design codes. In the Management of risks involving existing structures is
context of existing structures, it is particularly the direct responsibility of their owners, operators and
important to understand that design codes are occupiers. Structural engineers should take any
written for design, not for assessment. Use of a opportunities to recommend to those responsible
design code, within the scope of its application, that a strategy for management of the structure is

The Institution of Structural Engineers Risk in structural engineering 47

9.8 Risk management during the life of a structure

required and will be economically beneficial. This 2006-October 15, 2007. Report. Available at: http://
should cover records, maintenance and inspection www.cevc.gouv.qc.ca/UserFiles/File/Rapport/report_
and periodic review. When engaged on modification eng.pdf [Accessed: 6 September 2010]
work, and when refurbishment9.41 or repair is required,
structural engineers should follow appropriate parts of 9.13 Wood, J.G.M. ‘Implications of the collapse of the de la
the advice given in this Report for design and Concorde overpass’. The Structural Engineer, 86(1),
construction. 8 January 2008, pp16-18

9.14 Rail Accident Investigation Branch. Derailment of a

freight train near Stewarton, Ayrshire, 27 January
9.8 References 2009. Rail Accident Investigation Branch Report
02/2010. Available at: http://www.raib.gov.uk/cms_
resources.cfm?file=/100203_R022010_Stewarton.pdf
9.1 Neale B.S. ‘Maintaining structural safety through a [Accessed: 20 February 2013]
life-care plan and regulation’. Proceedings of the 3rd
Forensic Engineering Congress, San Diego, California, 9.15 Wood, J.G.M. Pipers Row Car Park, Wolverhampton:
September 2003. Reston, VA: ASCE Press, 2003, quantitative study of the causes of the partial collapse
pp433-440 on 20th March 1997. Chiddingfold: Structural Studies
& Design Ltd, 2002. Available at: http://www.hse.gov.
9.2 National Steering Committee for the Inspection of uk/research/misc/pipersrow.htm [Accessed:
Multi-storey Car Parks. Recommendations for the 20 February 2013]
inspection, maintenance and management of car park
structures. London: Thomas Telford, 2002 9.16 Standing Committee on Structural Safety. Confidential
reporting on structural safety for Scottish buildings.
9.3 Neale, B.S., ‘Better value and safety for refurbishment Livingston: Scottish Building Standards Agency, 2008.
projects through use of new standards’. Facilities Available at: http://cms.structural-safety.org/assets/
management and asset maintenance: applying and uploaded/documents/169_SCOTCROSS%20report.pdf
extending the global knowledge base: CIB W70 – [Accessed: 6 March 2013]
Proceedings, Glasgow, 18-20 September 2002. CIB
Publication 277 9.17 Fitzgerald, T.G. ‘Failure and rectification of fixings of
large pre-cast concrete cladding units’, Structural
9.4 Lloyd, C. ed. Asset management: whole life faults & repair ‘87: Proceedings, International
management of physical assets. London: Thomas conference on structural faults and repair held 7-9 July
Telford, 2010 1987, University of London. Vol 2. Edinburgh:
Engineering Technics Press. pp383-386
9.5 Neale B.S. ‘Hazard and risk assessments for
construction: a regulators view’. The Structural 9.18 Royles, R. ‘Repair of large cladding panels on multi-
Engineer, 73(22), 21 November 1995, pp388-90 storey structures’. Structural faults & repair ‘87.
Proceedings, International conference on structural
9.6 BS EN 1991-1-6:2005: Eurocode 1: Actions on faults and repair, University of London, 7-9 July 1987.
structures. Part 1-6: General actions – actions during Vol 2. Edinburgh: Engineering Technics Press.
execution. London: BSI, 2010 [Incorporating pp387-400
corrigendum July 2008]
9.19 Colford, B.R. and Clark, C.A. ‘Forth Road Bridge main
9.7 Neale, B.S., ‘Towards ensuring greater structural cables: replacement/augmentation study’. ICE
reliability through legislation’. In IABSE et al. Safety, Proceedings, Bridge Engineering, 163(BE2), June
risk and reliability: trends in engineering: conference 2010, pp79-89
report, international conference, Malta, March 21-23,
2001. Zurich: IABSE, 2001, pp759-764 9.20 Young, J. et al. ‘Assessment of the suspension cables
of the Severn Bridge, UK’. Creating and renewing
9.8 Office of the Deputy Prime Minister. The Building urban structures: tall buildings, bridges and
Regulations 2000 — Approved Document A: infrastructure: 17th Congress Report of IABSE,
Structure. 2004 edition incorporating 2010 Chicago, USA, 2008. Zurich: IABSE, 2008
amendments. Available at: http://www.planningportal.
gov.uk/buildingregulations/approveddocuments/parta/ 9.21 Cocksedge, C.P.E. and Bulmer, M.J. ‘Extending the life
documenta [Accessed: 20 February 2013] of the main cables of two major UK suspension
bridges through dehumidification’. Bridge Structures:
9.9 Building (Scotland) Regulations 2004. Edinburgh: The Assessment, Design and Construction, 5(4), 2009,
Stationary Office, 2004 (SSI 2004/406) pp159-172

9.10 BS EN 1991-1-7:2006: Eurocode 1: Actions on 9.22 Fennell, D. Investigation into the King’s Cross
structures – Part 1-7: General actions – Accidental underground fire. London: HMSO, 1988. Available at:
actions. London: BSI, 2010 [incorporating http://www.railwaysarchive.co.uk/documents/DoT_
corrigendum February 2010] KX1987.pdf [Accessed: 20 February 2013]

9.11 Neale B.S. ‘The Consequences of poor serviceability – 9.23 Health and Safety Executive, Collapse of a three-storey
and the way forward: keynote address’. In Byars, E.A. building: a report on the accident at Woodthorpe Road,
and McNulty, T. eds. Management of concrete Ashford, Middlesex that occurred on 1 August 1995.
structures for long-term serviceability. London: Thomas Sudbury: HSE Books, 1999
Telford, 1997, pp1-8
9.24 Neale B.S., ‘Forensic engineering in safety
9.12 Commission of Inquiry into the collapse of a portion of enforcement – some UK experiences’. In Srivastava,
the de la Concorde Overpass, October 3, N.K. ed. Structural engineering worldwide 1998:

48 The Institution of Structural Engineers Risk in structural engineering

 Risk management during the life of a structure 9.9

proceedings of the structural engineers world 9.40 BS EN 1991-1-4:2005+A1:2010: Eurocode 1: Actions

congress, San Francisco, 1998. Amsterdam: Elsevier, on structures – Part 1-4: General actions – wind
1998, paper T202-2 actions. London: BSI, 2011. [Incorporating
corrigenda July 2009 and January 2010]
9.25 Neale B.S., ‘Mitigation of failures due to inappropriate
loading during construction – a European code’. In 9.41 Neale B.S. ‘Assessment of structural safety risks’.
Neale, B.S. ed. Forensic Engineering – the Safety criteria for buildings and bridges: conference,
investigation of failures: proceedings of the 2nd Kensington, 1 July 1999. London: ICE, 1999, pp49-62
international conference, London, 12-13 November
2001. London: Thomas Telford, 2001, pp83-91

9.26 Ministry of Housing and Local Government. Report of 9.9 Bibliography

the Inquiry into the Collapse of Flats at Ronan Point,
Canning Town. London: HMSO. 1968
Actuarial Profession and Institution of Civil Engineers. Strategic
9.27 Fawcett, W. and Palmer, J. Good practice guidance on risk – a guide for directors. London: Thomas Telford, 2006
refurbishing occupied buildings. C621. London: CIRIA,
2004 ISO. TC 98/SC 2: Reliability of structures website. Available at:
http://www.iso.org/iso/standards_development/technical_
9.28 Neale, B.S.,’Teaching for enduring structural integrity’. committees/list_of_iso_technical_committees [Accessed:
In Allen, H.G. ed. Civil and structural engineering 18 February 2013]
education in the 21st century: proceedings of a
conference, 26-28 April 2000, University of
Southampton. Vol 2. Southampton: University of
Southampton, 2000, pp485-496

9.29 Vivian, S. et al. Climate change risks in building – an

introduction. C638. London: CIRIA, 2005

9.30 National Transportation Safety Board. Collapse of

I-35W Highway Bridge, Minneapolis,
Minnesota, August 1, 2007. Highway Accident Report
NTSB/HAR-08/03. Washington, DC: NTSB, 2008.
Available at: http://www.dot.state.mn.us/i35wbridge/
ntsb/finalreport.pdf [Accessed: 20 February 2013]

9.31 Railway Accident Investigation Unit. Malahide Viaduct

Collapse on the Dublin to Belfast Line, on the
21st August 2009. Investigation report no. R2010
004. Blackrock: RAIU, 2010. Available at: http://www.
raiu.ie/download/pdf/accident_malahide.pdf [Accessed:
20 February 2013]

9.32 Iddon, J. and Carpenter, J. Safe access for

maintenance and repair: guidance for designers.
C686. 2nd ed. London: CIRIA, 2009

9.33 Institution of Structural Engineers. The Operation and

maintenance of bridge access gantries and runways.
2nd ed. London: IStructE, 2008

9.34 ‘Gantry fall deaths ‘‘unlawful’’’. BBC News website.

21 July 2003. Available at: http://news.bbc.co.uk/1/hi/
england/3084859.stm [Accessed: 20 February 2013]

9.35 Institution of Structural Engineers. Appraisal of existing

structures. 3rd ed. London: IStructE, 2010

9.36 BS 8110: Structural use of concrete [in 3 parts]

9.37 BS EN 1992: Eurocode 2: Design of concrete

structures [in 4 parts]

9.38 CP3: Chapter V: Part 2: 1972: Code of basic data for

the design of buildings: Chapter V: Loading – Part 2:
Wind loads. London: BSI, 1972 [incorporating
amendments issued January, March and June 1986,
September 1988 and September 1993]

9.39 BS 6399-2:1997: Loading for buildings: part 2: code

of practice for wind loads London: BSI, 2002.
[incorporating amendment no.1 and corrigendum
no.1]

The Institution of Structural Engineers Risk in structural engineering 49

10 Risk management for demolition and refurbishment

10.1 Introduction examples of situations where hidden water

penetration has caused widespread corrosion. The
potential effects of any degradation on strength and
This chapter discusses risk management during and on the scale of refurbishment costs can be
in preparation for demolition10.1, partial demolition or formidable; it is a risk that has to be recognised and
structural refurbishment, giving guidance on the need assigned.
to understand the existing structure, managing
(deliberate) structural instability and reducing Risks as a result of working on, in or near to existing
uncertainty10.2, 10.3. Designing for future demolition is operational facilities are such that the risks of
covered in Section 7.9. unplanned events leading to personal injury incidents
can be higher than those on new build activities.
Risks to structure, people and surroundings should
be considered10.6.
10.2 The potential for unplanned events

Structures are often supported by a complex 10.3 The need to know the existing
interaction of structural mechanisms, not all of which
were intended by the designers. Arching, catenary structure
action and elements that were intended to be non-
loadbearing can provide load paths that were not
planned. Before any attempt is made to change or Risk management strategies for work on existing
demolish a structure, it is important to understand structures need to be as effective as and have a
what keeps it up. This requires an understanding of wider scope than those for new build projects. The
what is physically present (which is often hard to simple reason for this is that a thorough knowledge of
establish) plus a knowledge of how the components the structure to be worked upon, including the history
work together and what the load paths are to provide of use, is essential to minimise risk. This knowledge
overall stability. should be fed into assessments that lead to a
comprehensive understanding of structural behaviour,
Structures designed some time ago may have been both locally and throughout the structure, under any
designed on principles that are unfamiliar to today’s particular circumstances involving work that affects
designers. Before the days of computers, the the structure.
connection moments in steel frames were often
established using empirical rules. Steel being a ductile Assessments should take account of known faults in
material, some redistribution occurred under dead the structure, indications of potential faults and also
load and the initial live load, but for subsequent re- modifications undertaken during its life time. The aim
application of the live load the structure behaved is to avoid implementing inappropriate work on a
elastically. Although a fully elastic analysis might show structure that will affect structural integrity in adverse
an overstress, these structures may be quite ways and that may thus lead, for example, to
satisfactory and need no strengthening during premature collapse or flying debris.
refurbishment. The key is understanding the structure.
Since the existing records describing the structure
Structural refurbishment and demolition can be may not be completely accurate, any planned
journeys into the unknown, although they should not refurbishment strategy should proceed cautiously.
be. Older buildings often lack any drawings, let alone The structure should be probed and uncovered
‘as built’ records. Many structures have had so many sequentially and the project budgeting should allow
alterations and there is so much stability for the possibility of unearthing the unexpected. It is
interdependence within them that it is far from clear not unknown to uncover asbestos, which until the
what holds what up, so there is the ever present 1970s was used routinely without any awareness of
danger of removing a vital component. The classic its health hazards. In the UK there is now a
failure of the West Gate Bridge in Melbourne10.4 statutory duty to manage existing asbestos and
(1970) happened during construction, not keep a register of any that is present. A lack of
refurbishment, but it occurred because the erectors registered asbestos in a structure does not mean
removed certain bolts to ease construction, and, in so that none will be found; although it might change
doing, precipitated full scale collapse. This is not the legal liability, it will not protect the health of
unknown in demolition and refurbishment. anyone who finds some.

In 2010 CROSS reported on the demolition of two

13-storey large panel concrete tower blocks10.5. Even
though the demolition method was revised following 10.4 Structural refurbishment
each event, not less than four unexpected collapses
occurred. A major contribution to this was the lack of
adequate ties between the concrete panels. 10.4.1 Overview

Experienced engineers working on refurbishment are Refurbishment of existing structures may require
cautious about what might be found. There are many consideration of many of the same aspects as

50 The Institution of Structural Engineers Risk in structural engineering

 Risk management for demolition and refurbishment 10.5

design, but with the additional consideration that the In all cases where removal is planned and deliberate
structure will have aged, some details may be instability is to be introduced to remove material, an
unknown, and society’s expectations for a safe assessment for residual stability should be made. In
structure will have changed. Health issues also need other words, this should be considered a design
to be considered as well as environmental process. These assessments, and the associated
imperatives. work, should always be undertaken by suitably
competent people. Every structure is different; even
10.4.2 Structural stability those that may have been designed to be identical
will have been built differently and will have had
In many cases, any significant level of different histories. Co-ordination of temporary stability
refurbishment will mean that the structure has to is recommended for each occurrence of structural
comply with the regulations for new construction. work on a building, with the appointment made of
In the UK, this will mean, in particular, the someone in the role of a temporary stability
requirements against disproportionate collapse. coordinator.
Non-structural requirements, such as insulation
against sound and thermal loss, may also have an A flow chart showing considerations for structural
influence on the refurbishment process and stability is included in the British Standard code of
indirectly on the structure. practice for demolition (BS 6187:2011)10.9.

10.4.3 Fire precautions

Additional measures may be required for fire 10.6 Reducing uncertainty

compartmentation, fire stops and means of escape.
During refurbishment works, temporary fire protection
measures may be required. Demolition, including partial demolition, is a particular
case of the maxim that the effort put into managing a
10.4.4 Underpinning works risk should be proportionate to the consequences.
Demolition can involve significant uncertainty and a
Buildings are often extended by means of new conservative approach should always be taken.
basements, or require new foundations for Sometimes, it may be most effective to manage the
extension works. These may require underpinning risk by minimising the consequences; for example, by
of the existing foundations; many accidents have ensuring that nobody is in a position where they
occurred during underpinning. This requires the could be injured if anything unpredictable happens.
designer to have a good understanding of the
ground and the structure and in particular, requires A number of considerations specifically need to be
discipline and competence from those carrying out taken into account. These include:
the work on site. – Knowledge of the site.
– Decommissioning procedures, even for small
projects of any type.
– Structural hazards.
– Health hazards.
10.5 Managing (deliberate) structural – Protection of the environment, including managing
instability arisings and wastes.
– Health and safety of persons on or off site.
– Effects of dust (health, contamination, dust
An approach should be adopted such that the explosions).
structural integrity, or safety, is assessed for all – Safe working spaces and exclusion zones – even
stages of removal of any part of a structure. This for small jobs.
has become more important for modern structures – Principles of structural removal or demolition – for
that are highly efficient in the sense that every the structure concerned.
component plays a part in stability. Thus, erection – Avoidance of unplanned collapses.
of long steel rafters can be difficult since their – Temporary, or auxiliary, structures for stability and
design requires support of closely spaced lateral access.
purlins, back to a braced system. In reverse, – Demolition techniques – consider possible
removal of that bracing (or careless removal of the alternatives.
purlins) could initiate an unstable collapse. The – Materials handling and processing.
trend for modular building systems presents other – Completion of the works – objectives, achievement
problems since sometimes all members add to the and review.
structural stability. Understanding how a building
stands up is a prerequisite to understanding how Interaction of these issues is considered in
to take it down. Good examples of the skills BS 6187:201110.9.
required in demolition can be seen in the problems
of taking down post tensioned structures as, for
example, Marks & Spencer10.7 or Bernard
House10.8 in Manchester. 10.7 Guidance
It is a useful principle to assume there may be
structural instability everywhere whenever the A number of guides are available, as listed in the
structure is worked upon. This may include work that bibliography. Some of these give core advice,
may not (normally) be thought of as demolition while others depend on the particular
activities, or perhaps, even conform to all definitions circumstances that may apply; such as type,
of demolition, such as removal of material that is not height or location of the facility and the extent of
expected to affect the structural stability in any way. structure being removed.

The Institution of Structural Engineers Risk in structural engineering 51

10.8 Risk management for demolition and refurbishment

10.8 Conclusions and recommendations Bussell, M. et al. Retention of masonry façades – best practice
site handbook. C589. London: CIRIA, 2003

Demolition, whether complete or partial, involves the Clarke, R. ‘Role of the structural engineer in demolition’. The
intentional destabilising of a structure. Unlike Structural Engineer, 88(11), 2 June 2010, pp28-33
construction, the details of the structure and its
condition may not be fully known and it is important Neale, B. ‘Demolition, partial demolition, structural refurbishment
to ‘expect the unexpected’. Demolition therefore and decommissioning.’ In McAleenan, C. and Oloke, D. eds. ICE
requires structural engineers to have additional manual of health and safety in construction. London: Thomas
competence to deal with the risks involved. To help Telford, 2010, pp215-232
for the future, however, designers should ensure that
for their designs the structural principles and loading
options and criteria are available to their clients. This
should be with the recommendation that they are
kept, made available to those planning future works
and that they are passed onto successive owners
and operators, ensuring that they are updated to take
account of subsequent works.

10.9 References

10.1 Addis, W. and Schouten, J. Principles of design for

deconstruction to facilitate reuse and recycling, C607.
London: CIRIA, 2004

10.2 Marino Duffy B.M. et al. ‘Comparative study of UK and

Italian refurbishment sites involving demolition activities
and structural instability: risk factors and health & safety
management strategies’. Proceedings of the 2nd
International structural engineering and construction
conference (ISEC 02), Rome, 23-26 September 2003

10.3 Briggs, M. et al. Decommissioning, mothballing and

revamping. Rugby: IChemE, 1997

10.4 Report of the Royal Commission into the failure of the

West Gate Bridge. Melbourne, Victoria: Government
Printer, 1971

10.5 ‘Collapse of large panel structure buildings during

demolition’. CROSS Newsletter, 18, April 2010.
Available at: http://www.structural-safety.org/view-
report/cross273 [Accessed: 20 February 2013]

10.6 Egbu, C. et al. ‘Managing health & safety in

refurbishment projects involving demolition and
structural instability’, Facilities management and asset
maintenance: applying and extending the global
knowledge base: CIB W70 – Proceedings, Glasgow,
18-20 September 2002. CIB Publication 277

10.7 Roberts, J.M. ‘Demolition of Marks & Spencer,

Manchester (a six-storey commercial building
supported by post-tensioned beams)’. The Structural
Engineer, 77(2), 19 January 1999, pp20-25

10.8 Sellors, G., ‘Demolition of Bernard House, Piccadilly

Plaza, Manchester – January-July 2001’. The
Structural Engineer, 82(2), 20 January 2004, pp30-34

10.9 BS 6187:2011: Code of practice for full and partial

demolition. London: BSI, 2011

10.10 Bibliography

Bussell, M. et al. Retention of masonry façades – best practice

guide. C579. London: CIRIA, 2003

52 The Institution of Structural Engineers Risk in structural engineering

Appendix A Case studies

A.1 Introduction particularly in hurricanes. Severe damage can be

caused by localised wind-induced failure of parts of
buildings and there have been failures of temporary
Many of the concepts discussed in previous chapters structures where the judgement of short-term
may seem abstract and so might be better wind-loading has proved inaccurate.
understood by studying what has gone wrong in
practice. To illustrate this, and to put ideas into The predicted live loading in structures is
context, this chapter recounts some of the more fundamentally uncertain and can change with time, as
infamous failures and highlights lessons that might be a result of both better understanding and changing
learned. The chapter is arranged according to the conditions. Design code requirements for wind and
concepts illustrated in this Report. A key lesson is snow loading have changed significantly as further
that there have been, and continue to be, many research has been undertaken. Furthermore, climate
failures, many of them of high profile structures. With change is predicted to lead to further changes. Axle
hindsight many of the causes were predictable, so loads on bridges change with increases in the legal
why did they occur? The answer, it seems, is that it is limits for lorries and total bridge loading increases
painfully easy to overlook the obvious. The risks with traffic density. While not a failure in the sense of
certainly exist and part of risk reduction is to look for structural collapse, and not something that could
them. have been anticipated at the time of design, it would
have been better if the strengthening of the Severn
Learning from the past is a key activity, with the BridgeA.5 to withstand increased traffic loads could
purpose of minimising exposure to risk in the future. It have been avoided. An interesting question is: ‘To what
is worth reiterating that one of the purposes of extent should designers reduce the risk that
studying ‘near misses’ – sometimes referred to as modifications will be required, by allowing a margin in
‘near hits’ – is that, statistically, failures are rarely the original design?’
isolated incidents and they are a warning that action
might be required. The study of actual failures and The live loading defined in standards, such as
incidents demonstrates that many risks which might 4kN/m2 for office floor loading, only appears precise
be dismissed as ‘theoretical’ are actually real and because it is defined as such. The real loads are highly
deserve consideration. variable and, in practice, the design makes the
assumption that probably they will lie below the value
The case studies are not all about structural taken for design.
engineering. They are about the way humans have
managed or failed to manage risk, which is common
to all industries. They illustrate the importance of
understanding root causes, identifying the most
serious risks and learning from experience. While they A.3 Extension of technology to an invalid
are listed by principal cause, such events are rarely extent
due to just one cause and more usually disasters
result from the combination of technical and
procedural errors. The Tacoma Narrows Bridge failureA.6, A.7 (1940)
occurred primarily because the technology used
The information in this chapter is believed to be successfully on previous suspension bridges became
correct, although some interesting and well-known invalid on longer spans where aerodynamic effects
failures have had to be omitted since insufficient were disproportionately more important. This
information could be found to describe the events highlighted how much there was to learn about wind
without speculation. Some recent events have been aerodynamics and yet since that time several
omitted or given limited discussion as liability is still structures have suffered from oscillation in wind.
sub judice.
In 2006, the roof of the ice-arena in Bad Reichenhall,
The discussion of each failure is provided solely to Germany, collapsed under snow load. 15 people died
illustrate the principles of risk management and is not and 30 were seriously injured. The roof was
intended to represent the opinion of the authors or supported by special 48m long timber box girders.
the Institution in regard to the allocation of These were 2.87m deep, although the technical
responsibility or liability. approval for this type of girder limited the depth to
1.2m. A paperA.8 prepared for the public prosecutor
also identifies a number of other reasons that
contributed to the collapse. These were use of
A.2 Uncertainty in loading urea-formaldehyde glue under moist conditions, errors
in structural calculations (and failure to check them), a
lack of maintenance and vulnerability to progressive
The first Tay BridgeA.1, A.2 (1879) collapsed under wind collapse.
loading. While much has been learned since then, often
by bitter experience, failures due to wind loading or The long span box girder bridge was introduced over
wind-induced movement continue to occur, e.g., the a relatively short period around 1970, with little
dramatic failures of the Ferrybridge cooling opportunity for experience to be fed back into design.
towersA.3, A.4 (1965) or by more recent storm damage, The failures during construction at Milford HavenA.9

The Institution of Structural Engineers Risk in structural engineering 53

A.4 Case studies

(1970) and KoblenzA.10, A.11 (1972) were caused at A.5 Uncertainty in extreme loading
least partly by a failure to understand the loading
conditions during construction and the strength of
boxesA.12. A further issue was the relationship between Most of what is known about earthquake loading has
diaphragm buckling capacity and plate alignment with been derived from failure studies. As a result, the
respect to welding distortion imperfections. This history of seismic code development is one of
observation led to the Merrison InquiryA.13 (1973) and gradual increase in the lateral forces that buildings are
rules. Apart from the risk that the structure fails to work required to withstand, coupled with detailing rules to
as intended, the failure to anticipate tolerance demands avoid brittle failures observed in previous events.
(especially over material interfaces) is one of the Nevertheless, overall seismic forces remain
commonest causes of contractual disputes with fundamentally uncertain and, generally, designs are
associated delays. configured to address a series of seismic responses
based on different earthquake occurrence
These cases illustrate the difficulties of developing probabilities. Structures are configured to be ductile,
design into new areas. Structural engineering should the risk mitigation strategy being that the structures
develop if it is to serve society, but caution is required might deform more than expected but at least they
with structures which are essentially prototypes. won’t collapse. It is not usually economic to design
structures to resist earthquake forces and
displacement without damage, unless the function of
the structure requires this as, for example, in the case
A.4 Fatigue loading of nuclear reactors and hospitals in seismic zones.

Although not considered in the design of normal

An inability to recognise less common loading structures in the UK, both the Folkestone
conditions, particularly fatigue, has been the cause of earthquakeA.18 (2007) and the Birmingham tornadoA.19
many failures. The jack-up barge Sea Gem A.14 (2005) resulted in structural damage.
(1965), which found the first North Sea gas in UK
waters, collapsed and sank in the North Sea killing four
men. The Public Inquiry concluded that metal fatigue in
the system linking the hull to the legs was to blame. The A.6 Failure to understand materials
failure of the Alexander Keilland rigA.15 (1980) was also
due to fatigue and subsequent fracture. One of the
worst series of failures ever, the Comet aircraftA.16, was The limitations of construction materials are not
put down to metal fatigue. always apparent when they are first used. There have
been a series of scares on the long-term durability of
‘B of the Bang’ (2005) (see Figure A.1) was a concrete with degradation by sulphates, chlorides,
dramatic 56m tall sculpture, commissioned by alkali-silica reaction, thaumasite, mundic, etc. The
Manchester City Council to mark the 2002 collapse of roof beams in the Sir John Cass
Commonwealth Games. It consisted of 180 hollow SchoolA.20 (1973) highlighted the danger of high
spikes, made from weathering steel, radiating from alumina cement conversion and, in the ensuing
the top of an inclined support. Wind induced vibration nationwide investigation, many more cases of
resulted in the tips of several spikes becoming excessively weak structures were discovered. Similarly,
detached. Attempts to modify the dynamic response the failure of liberty ships in the Second World WarA.21
did not prevent further failures and the sculpture was showed the need for the engineering profession to
dismantled. The sculptor and the design and understand brittle fracture in steel.
construction subcontractors reached an out-of-court
settlement for £1.7 million with the city councilA.17. During construction of an elevated ramp to the Riley
Road InterchangeA.22, A.23 near Chicago (1982),
falsework to support cast-in-place concrete was
supported on precast plain concrete slabs about 1m2
and 300mm thick. Of the four defects found to have
contributed to the collapse, the principal cause was the
fracture of the unreinforced blocks during placement of
deck concrete. The resulting sudden change in
distribution of forces caused the falsework to move out
of line and collapse, killing 13 workers. This was the
worst construction accident in the history of Indiana.

The Malpasset Dam disasterA.24, A.25 (1959), which

killed more than 400 people, was at least partly caused
by an inadequate site investigation that had failed to
pick up certain key geological features. The Carsington
Dam failureA.26, A.27 (1984) was partly due to the failure
to foresee pre-existing shear surfaces.

A.7 Failure to identify the hazard

The failure to identify gas explosion as a hazard in the

Ronan Point flatsA.28 (1968) might be considered an
Figure A.1 ‘B of the Bang’ sculpture obvious omission, and in that case was compounded

54 The Institution of Structural Engineers Risk in structural engineering

 Case studies A.8

by inadequate ties between components. However, the Rock Ferry SchoolA.43 (1976) was a lesson in that,
explosion at the Abbeystead pumping stationA.29 while individual parts were designed correctly, the
(1984), which killed 16 visitors and injured many more, structure as a whole was unstable and the structure
was less predictable. The explosion was caused by the certainly lacked ‘robustness’. More tragically, the
ignition of a mixture of methane and air that had Aberfan coal waste slipA.44, A.45 (1966) killed 144
accumulated in the valve house after the methane had people (116 of them children) and was essentially a
seeped out of the tunnel walls. The designer, stability failure resulting from poor design and
constructor and operator were all initially found liable for maintenance; once movement occurred in the
damages. In a complex appeal, only the designer was saturated coal waste a huge volume surged down the
found liable. Two of the three law lords considered that slope. Likewise, the failure of Carsington DamA.26, A.27
the designer should have foreseen that methane could (1984) was a ground instability failure exacerbated by
accumulate in a void, while the third considered this uncertainties in ground conditions.
was not reasonably foreseeableA.30.
Incidents of crane overloading continue to occur,
There have been many incidents due to bridge either as a result of instability of the crane as a whole
impacts. In the UK, between 1994 and 2001, there or instability of some part such as the jib. Most of
were an average of 1500 incidents per year of road these relate to misuse of the crane, such as
vehicles striking railway bridgesA.31. Bridge collapses overloading or incorrect erection, or failure to ensure
also occur following scour by river currents, for adequate support to outriggers.
example, the Malahide Viaduct in IrelandA.32, A.33, A.34.

Some of the worst cases of bridge failures have

occurred following impact damage to the piers by A.10 Errors in design or detailing
shipping. A vessel collision with an Amtrak bridge in
AlabamaA.35 (1993) cost 47 lives and millions of dollars.
An incident with a large ship took out a complete section One of the causes of the Ramsgate link-bridge
of the Tasman Bridge in Hobart, causing immense collapseA.46, A.47, A.48, (1994) was a design/detailing
disruptionA.36, A.37 (1975). It is now normal practice to error made by the designer and missed by the
provide barriers to prevent ships getting too close to checking organisation. This might not have been fatal
bridge piers, but even then the protection can only be had the bridge been provided with some form of
designed for the largest anticipated vessel. The Erskine redundancy; as it was, failure of one bearing led
Bridge over the Clyde in Glasgow was severely inevitably to complete collapse. The designer, checker
damagedA.38 (1996) by the simple error of sending an oil and client were all convicted subsequently. More details
rig underneath whose mast was too tall for the clearance of this case are given in The Structural EngineerA.49.
at that time in the tidal cycle. The damage cost £3.7m to
repair with a further £0.7m in lost revenue due to closure. Cases have been reported of designs where
reinforcement meshes have been simply transposed,
or inadequately placed on site, leading to collapse or
necessitating demolition. A car park in Birmingham,
A.8 Errors in dynamics UK required external reinforcement to the slab soffits
when it was discovered, some time after completion,
that the tension reinforcement required for the short
The Tacoma Narrows failure (see Section A.3) is a span had been detailed for the long span and vice
clear example. The dramatic failure of the 365m high versa.
Emley Moor TV mastA.39 (1969) was another. The mast
consisted of a tubular steel section up to 275m with a
lattice section above. Failure occurred in strong winds
with large quantities of ice formed on the tower and its A.11 Deterioration and lack of
guys. The cause was held to be dynamic oscillation, maintenance
rather than the direct effects of the weight of ice, and at
that time the phenomenon was not widely understood.
This was an inadequate defence for the designers in A gantry girder bridge over the railway collapsed at
subsequent litigation on liability, not resolved until 1983, ClaphamA.50 (1965) due to the build-up of rust in
with a settlement of £3.2 million. overlapping plates bursting the riveted seams apart.
The car park failure at Pipers Row Car ParkA.51 (1997)
More problems occurred on the Millennium was probably exacerbated by top-mat reinforcement
Footbridge in LondonA.40, A.41 (2000), which swayed corrosion. A waitress was killed in EdinburghA.52,
excessively under crowd loading and it was necessary (2000) by falling masonry and this highlighted a more
to backfit the bridge with dampers. The problems with widespread problem leading to a reportA.53 on falling
the dynamic response of football stands under crowd masonry. As a result CROSS was commissioned by the
loading are equally important. There has been a long Scottish Building Standards Agency to gather data and
history of stand failures such as that at the Pink Floyd report findingsA.54, A.55.
concert at Earl’s Court, LondonA.42 (1994), and the
worst cases have been in temporary stands where A number of major bridges in the UK have had to
some have collapsed with loss of life. undergo substantial repairs that have caused
significant, but probably unattributed, economic
disruption. Two examples are the Kingston Bridge in
GlasgowA.56 and the Thelwall Viaduct on the M6
A.9 Errors in stability motorwayA.57 which were partially closed for several
years while repairs and improvements were made.
These were necessitated partly by a failure to anticipate
There are examples of error in stability in both an increase in traffic levels, resulting in the need to
buildings and in the ground. The collapse of part of increase the load capacity, but mainly as a result of

The Institution of Structural Engineers Risk in structural engineering 55

A.12 Case studies

defects. In the case of the Thelwall Viaduct in Cheshire, killed and eight injured. The building used flat slab
the bearings failed only a few years after being replaced construction and progressive failure occurred during
as part of a complete re-build of the bridge deck. This demolition. Punching shear failure could be seen in the
failure was compounded by the lack of provision for slab. The main cause of failure was excessive debris
replacement, meaning that the deck had to be and other loads on slabs.
strengthened before it could be jacked up to access
the failed bearings.

A.14 Human factors

A.12 Identifying significant risks
The Moorgate underground train crashA.65, A.66 (1975)
killed 43 people and injured 70 more. The cause was
Ten people were killed when a Land Rover towing a never established, although it was clear that the driver
trailer went off the M62 motorway at Great Heck near had made no attempt to slow the train. It illustrated
Selby, UKA.58 (2001). The vehicle plunged on to the graphically the need to consider human factors in
railway line into the path of a passenger train, which design. Subsequently, engineered systems were put in
derailed and impacted with a goods train travelling in place at ‘dead-end’ tunnels which would stop trains
the opposite direction. Some people at the time automatically. Other train crashes have been linked to
described it as ‘a chance in a million’. Considering, drivers missing red signals, e.g. Ladbroke
however, the number of bridges there are over railways GroveA.67, A.68 (1999).
and the statistics regarding the number of vehicles
which run off the road, the probability of a vehicle going The significance of this case is that it may not be
off the road at a railway bridge at some time is not low. adequate to rely on one person taking the right action
At this particular site, once that had happened, the when the consequence of failure is high or there is a
probability of ending up on the line was high. risk of malicious action, e.g. terrorism. If the hazard
Thereafter, given rail traffic density, once the vehicle cannot be engineered out, the management
was on the track the probability of being hit by a train arrangements may need to include supervision.
was also high. So the total probability was not small
and the potential consequence was significant.

On the other hand, a tornado seriously damaged A.15 Design change

several houses in LondonA.59 (2006); yet engineers do
not customarily design for tornado damage in the UK
because the probability of any individual property being Two aerial walkways collapsed at the Kansas City
damaged is small and the cost of protecting all houses Hyatt RegencyA.69 (1981), killing 110 people and
would not be justified. For such limited risks, insurance injuring more than 200. The immediate cause of the
can limit the financial risk to each householder to the collapse was a poorly fabricated connection between
average for the nation. the walkway’s supporting steel rod and the walkway
beams (see Figure A.2). The fabricator had changed
The fire at the Bradford Football Club standA.60, A.61 the design to simplify assembly, inadvertently doubling
(1985) spread rapidly causing loss of life, partly the load on the connector. In addition, the cross beams
because the crowd could not escape onto the pitch. where the hanger bars anchored should have been
The fire highlighted the balance of risk assessment made of two rolled steel channels (RSCs) back to back
needed; on the one hand in keeping crowds from (with the hanger rod passing between them). However,
spilling onto the pitch to preserve public order versus for aesthetic reasons, two RSCs were welded to form a
the hazard of preventing them from being able to box section, introducing large local bending stresses in
escape in times of dire necessity. the flanges of the RSCs. Ultimate load tests of a similar
assembly showed that the detail was only strong
enough to take 10% of the design load. Responsibility
was disputed because of design changes and
A.13 Demolition uncertain communications.

The collapse of a concrete canopy during alteration

A building in Bootle, Merseyside, collapsed during works occurred at Albert House, Aberdeen, Hong
demolitionA.62, A.63 (2000). The site was a terrace of KongA.70 (1994) resulting in one death and several
nine three-storey Victorian properties, with basement injuries. Evidence pointed to lack of maintenance, plus
cellars. The building itself was somewhere between 90 unauthorised works resulting in extra loadings on the
and 100 years of age. One worker was buried under canopy.
rubble and died. In the opinion of the investigators, the
significant causes of the collapse were alterations to
the structural form of the original building and the
weakened state of the structure. Both were A.16 Temporary works and construction
foreseeable and could have been detected by a failures
structural survey if there had been one. The actual
demolition activity itself left the party wall with little or no
lateral support. There were also deficiencies in There have been many temporary works and
information given to the demolition contractor and the construction failures. The Barton High Level Bridge
planning of the site work (for which insufficient time had collapsesA.71, A.72 (1959) were failures of temporary
been allowed). works and stability.

The collapse of an industrial building at Yau Tong, The 68m high falsework collapse at Almuñécar in
Hong KongA.64 (2001) resulted in six people being SpainA.73 (2005) is an example of a temporary works

56 The Institution of Structural Engineers Risk in structural engineering

 Case studies A.17

failure with serious consequences. According to a

report from the judicial inquiry (April 2007), the
collapse was caused by a defective welding seam
fracturing due to excessive stress caused by a
defective bolt. For such a small defect to be so critical,
however, suggests that robustness may have been
lacking as well.

A.17 Inadequate procedures

Disasters as diverse as the Kings Cross Underground

FireA.74 (1987), Piper AlphaA.75, A.76 (1988) and
HillsboroughA.77 (1989) all highlighted the need to
have adequate management procedures for safety, Figure A.2 Damaged channels at Hyatt Regency
as in each case loss of life was made worse by failures
to plan in advance a response in the event of a
disaster.
originally designed for low-rise construction in a
The football stadium incident at Hillsborough led to a country without a tradition of piped gas supplies. The
renewed edition of Guide to safety at sports transposition to the UK took no account of this.
groundsA.78. This was first published in 1975 and now Moreover, quality control was poor and the
collectively takes account of a number of sports ground investigation discovered that even the nominal
incidents. connections were incomplete. These issues might
have been addressed if the severity of the potential
consequences had been considered. Nobody seems
to have asked ‘‘what’s the worst thing that can
A.18 Systems failures happen?’’

The attack on the Pentagon in Washington DCA.82

Civil engineering structures are commonly required to (2001), using a hijacked airliner, caused the death of
interact with other engineered systems. They are 125 people on the ground; this was, however, far fewer
frequently called upon to enhance safety by acting as than might have been the case. The airliner impacted
engineered barriers to assist in providing plant safety. the lower floors of the Pentagon building, causing
The apparent failure of the control system at extensive damage, but the continuous spiral links to the
BuncefieldA.79 (2005) led to the biggest deflagration columns confined the core concrete and allowed it to
and fire in Western Europe since the Second World maintain some strength. In combination with catenary
War (see Figure A.3). Good examples of the chain of action in the continuously reinforced floor beams, this
events that may initiate a disaster can be taken from limited the area of collapse of the upper floors and
other engineering disciplines. The failure to check that allowed their occupants to escape (See Figure A.4;
the bow door was closed, plus a lack of safety note that use of military imagery does not imply or
management, on the Herald of Free EnterpriseA.80 constitute endorsement of the Institution of Structural
(1987) led to the car deck being flooded in a manner Engineers, its products, or services by the U.S.
that made the ship unstable and led to it capsizing. The Department of Defense.)
failure of a tyre on ConcordeA.81 (2000) generated
debris that penetrated the plane’s fuel tanks which,
coupled with the design of the tanks, led to the crash
and subsequent fleet abandonment. Lessons from
such failures are used to develop safety thinking in
areas of complex interaction such as nuclear and
chemical plant design.

A.19 Robustness

The collapse of the 22-storey Ronan PointA.28

apartment tower, London (1968) was initiated by a
relatively small explosion. Only two months after
completion an accidental gas explosion on the 18th
floor blew out an external wall panel, triggering
progressive collapse of the whole corner of the
building. The incident was the main driver for the
introduction of robustness rules into the UK Building
Regulations.

The building was built using a ‘large panel system’ of

precast concrete panels for walls and floors. These
relied on gravity for stability, with only nominal
connections between them. It was a Danish system, Figure A.3 Warehouse damaged by blast at Buncefield

The Institution of Structural Engineers Risk in structural engineering 57

A.20 Case studies

Figure A.4 Damage to the Pentagon

A.20 Mobile structures A.21 Failure to learn from previous incidents

The Severn Bridge gantry accidentA.83 (1990) In 1994, tunnels were being driven at Heathrow
occurred while moving the gantry from its station to Airport, UK, using the ‘New Austrian Tunnelling
its position of work. It was designed concurrently Method’A.86 (NATM). This is an ‘observational’ method,
with the bridge deck to allow inspection and in which feedback from the construction process is
maintenance of the permanent works (steel box used to modify the approach, if necessary, as work
girder), and was being used by a contractor to paint progresses. NATM takes advantage of the ability of the
the box girder. The gantry geometry was intended to ground to support itself, at least partly, if cut in an
have been locked but previous misuse had left it in an appropriate way. Sprayed concrete is then used to
unlocked mode. form a permanent support. The process is dependent
on the designer having a good understanding of the
The Avonmouth Bridge accidentA.84 (1999) involved ground conditions and a skilled, experienced and
a construction gantry travelling on runway beams appropriately motivated site team.
designed for a permanent works inspection gantry.
Strengthening works to the bridge superstructure The face of the Heathrow tunnel collapsed during a
and replacement of the existing permanent gantry night shift, resulting in further collapses over three
runway beams were combined into a single days and subsidence of the airport above the
operation. The front hangers were running on old tunnelA.87, A.88. There were no fatalities but the airport
runway beams with the rear hangers suspended suffered severe disruption and the construction
from sections of newly installed runway beams. At programme was seriously delayed. An adjacent rail
the time of the accident there was a gap of tunnel, in use, came close to being affected, and work
approximately 3.5m between the ends of the new was stopped on tunnels in central London which used
and the end of one of the old runway beams. Wind a similar method. The principal contractor was
blew the gantry along the rails and it rolled through subsequently fined £1.2 million.
the gap, falling down and killing four construction
workers. The cause of the accident was At the time of the collapse, information existed on
unsatisfactory hardware and unsafe working 116 previous collapses related to NATM. Kevin
practices, ineffective restraints on beams to prevent Myers, the HSE’s chief inspector of construction, said
longitudinal movement of gantries and no provision that the collapses could have been prevented but for
of end stops as a ‘fail-safe’ in the event of a cultural mindset which focused attention on the
restraints not working effectively. Above all, there apparent economies and the need for production
was failure to anticipate the mode of failure; to think rather than the particular risks.
‘what can possibly go wrong?’ The underlying cause
was documented as a general failure to plan,
organise, control, monitor and review the
operations; in particular, the lessons should have A.22 Safety culture
been learned from the earlier incident. Further advice
on the design of bridge access gantries is
contained in the Institution of Structural Engineers’ The space shuttle, Challenger, exploded shortly after
reportA.85. launchA.89 (1986). The primary cause was the failure of

58 The Institution of Structural Engineers Risk in structural engineering

 Case studies A.23

an O-ring seal made brittle by cold weather, which In this respect, the safety of all structures is
allowed hot gas to impinge on the external fuel tank underpinned by the quality assurance (QA) regimes
and its supports, leading to rupture of the fuel tank, a required for assurance of product quality. Long
fireball and disintegration of the craft. This was a known supply chains provide an opportunity for low quality
problem and a workaround existed, related to the products with forged certification to be supplied; in
temperature before launch, but those who decided to many cases the final purchaser can only distinguish
launch the shuttle did not fully understand the danger. these from the specified product by carrying out
The lessons for safety culture in NASA were not fully tests. While counterfeit components have not, so far
learned, leading to the loss of Columbia from a different as the authors are aware, resulted in serious
physical cause 17 years laterA.90. structural failures, they have been responsible for loss
of aircraftA.95.
The physical root cause of the loss of Columbia
(2003) was impact of a piece of insulating foam onto
a heat resisting tile. The official enquiryA.91 concluded,
‘‘that NASA’s organizational culture had as much to do A.24 Failure to understand the structure
with this accident as foam did’’. NASA’s formal safety
policies, although outwardly making safety the top
priority, were not fully effective in practice. The Navier’s suspension bridgeA.96 crossing the Seine at
independent safety reviewers were funded by the Les Invalides, Paris (1826), was almost completed
project they reviewed, resulting in a conflict of interest. when cracks in the foundations provoked the need for
remedial work which had significant financial, political
Lessons from the loss of Challenger had been and engineering consequences. The case was high
addressed but not absorbed into the culture. When profile; the project was declared to be a matter of
engineers raised concerns which had schedule or national pride with no economic motivation but,
cost implications, they were put down or ignored by surprisingly, was financed privately.
managers. A number of necessary improvements to
safety were identified, but were later cancelled on At the time the mathematics of the hanging chain
cost grounds. There was a view that normal flight were being elegantly developed to the satisfaction of
should continue, unless the shuttle was proved to be those involved. Navier developed and applied these
unsafe, rather than an expectation that every activity equations to complement the empirical experience of
should first be shown to be safe. Foam struck the Brunel, in Bristol, UK. The superstructure was
shuttle on a number of missions, but it survived with therefore designed from theoretical considerations.
limited damage, which led to a belief that it would The masonry foundations, however, were designed
always survive, irrespective of the engineering issues by traditional empirical methods.
and safety margins. All these factors created an
environment within which potentially serious The bridge experienced unexpected cracking in the
engineering issues could grow and become actual life foundations which anchored the main suspension
threatening problems, culminating in the loss of the cables. The cracking first occurred as the permanent
shuttle. load of the deck was applied to the ties, then more
extensively when a broken water main flooded one
This is also an example of the value of studying ‘near foundation. The combination of the increased pore
misses’. It was not the first time foam had become pressure, the strain compatibility in the curved
detached; on Concorde it was not the first time tyre masonry compression pier and the lower tie segment
damage had occurred and the Kings Cross escalator and friction restraining slipping on the saddle will all
was not the first wooden escalator to have caught fire. have contributed in different ways to the cracking.

Interfaces, such as between the iron superstructure

and the masonry infrastructure (one theoretical, one
A.23 Competence and quality empirical) are often imperfectly resolved.

In Coco Beach, Florida, a four-storey building

collapsed during constructionA.92. Investigation A.25 Novel design
disclosed that the building had not been designed for
shear capacity at the slab to column connection. The
severely overloaded connection collapsed, killing The failure of the roof of a departure hall at Roissy
several workers; it was subsequently determined that Terminal 2E airport (Charles de Gaulle)A.97 by
the engineers formally responsible for the design were punching shear, a year after commissioning, caused
aeronautical rather than structural engineers. several deaths. Although tragic, the failure is interesting
in terms of risk because it appears to have been a result
In Christchurch, New Zealand, the Canterbury of both engineering errors and organisational faults.
Television building collapsed in an earthquake (2011). A fundamental question is ‘how could such a prestige
The Royal Commission of Inquiry into the Building project with such a skilled design team get it wrong?’
Failure Caused by the Canterbury Earthquakes has There are numerous lessons.
not yet issued its final report on this building, but the
inquiry gave significant consideration to whether the There was an initial error in the design concept,
designers of the building had appropriate leading to cracking which was known about but not
expertiseA.93 and it has been alleged that the engineer fully responded to. Failure occurred by punching
who supervised its construction had fraudulent shear of the radial steel spacer struts, through the
credentialsA.94. concrete shell, which had insufficient steel
reinforcement. A prototype had demonstrated that
There are increasing concerns over sub-standard the design worked, that the details could be made,
building components such as cements and bolts. the concrete could be placed and that it stood up.

The Institution of Structural Engineers Risk in structural engineering 59

A.26 Case studies

The inadequately reinforced concrete resisted the A.9 Shirley-Smith, H. Report on collapse of Milford Haven
punching shear in the prototypes, and in the first year Bridge: fatal accident, 2nd June 1970. [s.l.]: [s.n.],
of use, by a combination of arching and tension in 1970
the concrete, but the cyclical temperature stresses,
time and progressive cracking reduced the resistance A.10 ‘Koblenz report pinpoints collapse cause’. New Civil
to punching. The structural concept was not Engineer, May 1972, p10
intrinsically robust, but individual modules were
nominally linked together to obtain some mutual A.11 Cottrill, A. ‘Stability theory blamed for Koblenz bridge
support. The linking, however, not only failed to collapse. New Civil Engineer, 23 November 1972,
prevent collapse but allowed one module to drag pp10-11
down the adjacent one.
A.12 Institution of Civil Engineers. Steel box girder bridges:
Finally, although there were three layers of procedures proceedings of the international conference, London,
in place for checking, these seem to have been too 13-14 February 1973. London: ICE, 1973
procedural, stifling the creative thought that is
required for true assessment of risk. One issue which A.13 Department of the Environment et al. Inquiry into the
may need to be addressed is that much of the basis of design and method of erection of steel box
post-accident analysis has been carried out for the girder bridges: report of the [Merrison] Committee.
benefit of the insurers, and has not been made London: HMSO, 1973
accessible for the engineering community to learn the
lessons. A.14 Ministry of Power. Report of the Inquiry into the causes
of the accident to the drilling rig Sea Gem. London:
The lack of redundancy in the operation of the HMSO, 1967
building was more expensive for the airport than
rebuilding the structure; the failure of one bay closed A.15 The Alexander L. Kielland accident: report of a
the whole terminal for several years. The extensive Norwegian public commission appointed by royal
repetition of a flawed design amplified the decree of March 28, 1980, presented to the Ministry
consequences of the error. The judicial inquiry as to of Justice and Police. [s.l.]: [s.n.], 1981
the causes and responsibilities of the tragedy of
Roissy remained ongoing in 2013. A.16 Withey, P.A. ‘Fatigue failure of the De Havilland Comet
1 Engineering’. Engineering Failure Analysis, 4(2), June
1997, pp147-154

A.26 References A.17 ‘Manchester B of the Bang sculpture core sold for
scrap’. BBC News website, 4 July 2012. Available at:
http://www.bbc.co.uk/news/uk-england-manchester-
A.1 Martin, T. and MacLeod, I.A. ‘The Tay Bridge disaster: 18703854 [Accessed: 25 February 2013]
a reappraisal based on modern methods of analysis’.
ICE Proceedings, Civil Engineering, 108(2), May 1995, A.18 British Geological Survey. Folkestone Earthquake 28
pp77-83 April 2007 07:18 UTC (08:18 BST) 4.2 ML. Available
at: http://www.earthquakes.bgs.ac.uk/education/
A.2 Thomas, J. The Tay Bridge disaster: new light on the reports/folkestone/folkestone_28_april_2007.htm
1879 tragedy. Newton Abbot: David & Charles, 1972 [Accessed: 25 February 2013]

A.3 CEGB. Report of the Committee of Inquiry into collapse A.19 ‘Tornado injures 19 in Birmingham’. BBC News
of cooling towers at Ferrybridge, Monday, 1 November, website, 28 July 2005. Available at: http://news.bbc.
1965. London: CEGB, 1966 co.uk/1/hi/england/west_midlands/4725279.stm
[Accessed: 25 February 2013]
A.4 Institution of Civil Engineers. Natural draught cooling
towers – Ferrybridge and after: proceedings of a A.20 Bate, S.C.C. Report on the failure of roof beams at Sir
conference, London, 12 June 1967. London: ICE, John Cass School Stepney. BRE Current Paper 58/74.
1967 Garston: BRE, 1974

A.5 ‘Severn Bridge and Wye Bridge Strengthening Work A.21 Failure Knowledge Database/100 selected cases: Brittle
1985-1991’. Available at: http://www.severnbridge.co. fracture of Liberty Ships. Available at: http://www.
uk/history_strengthening.shtml [Accessed: 25 February sozogaku.com/fkd/en/hfen/HB1011020.pdf [Accessed:
2013] 25 February 2013]

A.6 Walshe, D.E. and Wyatt, T.A. ‘Bridge aerodynamics: A.22 Carino, N.J. et al. ‘Investigation of East Chicago ramp
50 years after Tacoma Narrows’. Journal of Wind collapse’. ASCE Journal of Construction Engineering
Engineering and Industrial Aerodynamics, 40(3), 1992, and Management, 110(1), March 1984, pp1-18
pp317-336
A.23 Carino, N.J. et al. Investigation of construction failure
A.7 American Society of Civil Engineers. Tacoma Narrows of the Riley Road interchange ramp, East Chicago,
Bridge failure: report to the Committee of the Board of Indiana. NBSIR 82-2583. Available at: http://www.nist.
Direction. New York: ASCE, 1942 gov/manuscript-publication-search.cfm?pub_id=
908826 [Accessed: 25 February 2013]
A.8 Winter, S. and Kreuzinger, H. ‘The Bad Reichenhall ice-
arena collapse and the necessary consequences for A.24 Londe, P. ‘The Malpasset Dam failure’. Engineering
wide span timber structures’. World conference on Geology, 24(1-4), 1987, pp295-329
timber engineering. Miyazaki, Japan, 2008, paper
271. Available at: http://www.ewpa.com/Archive/2008/ A.25 Maurenbrecher, P.M. ‘The Malpasset Dam failure
june/Paper_271.pdf [Accessed: 25 February 2013] investigation and analysis examined’. In Neale, B.S. ed.

60 The Institution of Structural Engineers Risk in structural engineering

 Case studies A.26

Forensic engineering: from failure to understanding. A.39 Smith, B.W. ‘Case studies: Emley Moor mast’. In
London: Thomas Telford, 2009, pp229-238 Campbell, P. ed. Learning from construction failures:
applied forensic engineering. Latheronwheel: Whittles,
A.26 Skempton, A.W. and Vaughan, P.R. ‘The Failure of 2001, pp274-277
Carsington Dam’. Géotechnique, 43(1), March 1993,
pp151-173 A.40 Arup. The Millennium Bridge. Available at: http://www.
arup.com/MillenniumBridge [Accessed: 25 February
A.27 Coxon, R.E. Failure of Carsington embankment: a 2013]
report to the Secretary of State for the Environment.
London: HMSO, 1986 A.41 Dallard, P. et al. ‘The London Millennium footbridge’.
The Structural Engineer, 79(22), 20 November 2001,
A.28 Ministry of Housing and Local Government. Report of pp17-33
the Inquiry into the collapse of flats at Ronan Point,
Canning Town. London: HMSO. 1968 A.42 Penman, D. ‘Pink Floyd ‘very angry and upset’ over
accident: human error could have caused temporary
A.29 Health and Safety Executive. The Abbeystead stand’s collapse at rock concert attended by 15000
explosion: a report of the investigation by the fans’. The Independent, 14 October 1994. Available at:
Health and Safety Executive into the explosion on http://www.independent.co.uk/news/uk/pink-floyd-very-
23 May 1984 at the valve house of the Lune/Wyre angry-and-upset-over-accident-human-error-could-
water transfer scheme at Abbeystead. London: have-caused-temporary-stands-collapse-at-rock-
HMSO, 1985 concert-attended-by-15000-fans-danny-penman-
reports-1442784.html [Accessed: 25 February 2013]
A.30 Eckersley & Others v. Binnie & Others. Court of Appeal
(Civ Div). Fox, Bingham and Russell L.JJ http:// A.43 Menzies, J.B. and Grainger, G.D. Report on the
oxcheps.new.ox.ac.uk/new/casebook/cases/Cases collapse of the sports hall at Rock Ferry
Chapter 13/Eckersley v Binnie.doc [Accessed: Comprehensive School Birkenhead. BRE Current Paper
25 February 2013] 69/76. Garston: BRE, 1976

A.31 Hudson, S. Bridge Strikes – Special Topic Report. A.44 Report of the Tribunal appointed to inquire into the
London: Railway Safety, 2002. Available at: disaster at Aberfan in October 1966. London: HMSO,
http://www.rssb.co.uk/sitecollectiondocuments/pdf/ 1967
reports/Bridge Strikes – Special Topic Report.pdf
[Accessed: 25 February 2013] A.45 Austin, J. Aberfan: the story of disaster. London:
Hutchinson, 1967
A.32 Institution of Structural Engineers. Guide to
inspection of underwater structures. London: IStructE, A.46 Health and Safety Executive. Walkway collapse at Port
2001 Ramsgate: a report on the investigation into the
walkway collapse at Port Ramsgate on 14 September
A.33 Railway Accident Investigation Unit. Malahide Viaduct 1994. Sudbury: HSE Books, 2000
Collapse on the Dublin to Belfast Line, on the
21st August 2009. Investigation report no. R2010 A.47 Barber, J. ‘Ramsgate walkway collapse: legal
004. Blackrock: RAIU, 2010. Available at: http://www. ramifications’. In Neale, B.S. ed. Forensic engineering:
raiu.ie/download/pdf/accident_malahide.pdf [Accessed: a professional approach to investigation. London:
25 February 2013] Thomas Telford, 1998, pp39-60

A.34 Department of Transport. Report on the collapse of A.48 Fleming, D. ‘Design doubts: why the Ramsgate walkway
Glanrhyd Bridge on 19th October 1987. London: collapsed’. New Civil Engineer, 22 September 1994
HMSO, 1990. http://www.railwaysarchive.co.uk/
documents/DoT_Glanrhyd1987.pdf [Accessed: A.49 Chapman, J.C. ‘Collapse of the Ramsgate Walkway’.
25 February 2013] The Structural Engineer, 76(1), 6 January 1998,
pp1-10
A.35 National Transportation Safety Board. Derailment of
Amtrak Train No. 2 on the CSXT Big Bayou Canot A.50 Ministry of Transport. Enquiry report on the failure of
Bridge near Mobile, Alabama, September 22, 1993. an overbridge that occurred on 10th May 1965 at
RAR-94/01. Summary available at: http://www.ntsb. Clapham Junction in the Southern Region, British
gov/publictn/1994/RAR9401.htm [Accessed: Railways. London: HMSO, 1965. Available at:
25 February 2013] http://www.railwaysarchive.co.uk/documents/MoT_
Clapham1965.pdf [Accessed: 25 February 2013]
A.36 ‘Bridge gone: the severed artery’. Available at: http://
www.parliament.tas.gov.au/history/brd1.htm. A.51 Wood, J.G.M. Pipers Row Car Park, Wolverhampton:
[Accessed: 25 February 2013] quantitative study of the causes of the partial collapse
on 20th March 1997. Chiddingfold: Structural Studies
A.37 Court of Marine Enquiry. S.S. Lake Illawara. Report & Design Ltd, 2002. Available at: http://www.hse.gov.
157, Canberra: Australian Government Publications uk/research/misc/pipersrow.htm [Accessed:
Service, 1975 25 February 2013]

A.38 Tinning, W. and Watson, C. ‘Erskine Bridge is closed A.52 ‘Waitress killed by falling masonry’. BBC News
after oil-rig collision towing accident ends in severe website, 29 June 2000. Available at: http://news.bbc.
damage’. The Herald, 5 August 1996. Available at: co.uk/1/hi/scotland/811988.stm [Accessed:
http://www.heraldscotland.com/sport/spl/aberdeen/ 25 February 2013]
erskine-bridge-is-closed-after-oil-rig-collision-
towing-accident-ends-in-severe-damage-1.442658 A.53 Construction Industry Council for Scotland. Risks to
[Accessed: 25 February 2013] public safety from falling masonry and other materials:

The Institution of Structural Engineers Risk in structural engineering 61

A.26 Case studies

report of skills and training working party. [s.l.]: [s.n.], http://www.railwaysarchive.co.uk/documents/HSE_Lad_

2006 Cullen001.pdf [Accessed: 25 February 2013]

A.54 Standing Committee on Structural Safety. SCOTCROSS A.68 Lord Cullen. The Ladbroke Grove Rail Inquiry. Part 2
report. Livingston: Scottish Building Standards Agency, report. Sudbury: HSE Books, 2001. Available at: http://
2006. Available at: http://www.scotland.gov.uk/ www.railwaysarchive.co.uk/documents/HSE_Lad_
Resource/Doc/217736/0090815.pdf [Accessed: 6 Cullen002.pdf [Accessed: 25 February 2013]
March 2013]
A.69 Marshall, R.D. et al. Investigation into the Kansas City
A.55 Standing Committee on Structural Safety. Confidential Hyatt Regency walkway collapse. NBS Building Science
reporting on structural safety for Scottish buildings. Series 143. Washington, D.C.: NBS, 1982
Livingston: Scottish Building Standards Agency, 2008.
Available at: http://cms.structural-safety.org/assets/ A.70 Aberdeen Winner Investment Co Ltd v Incorporated
uploaded/documents/169_SCOTCROSS%20report.pdf owners of Albert House [2004] 3 HKLRD 910
[Accessed: 6 March 2013]
A.71 Merchant, W. ‘Three structural failures: case notes and
A.56 Brodie, A. and Telford, I. ‘Kingston Bridge general comments’. ICE Proceedings, 36, March
strengthening: phase 1’. The Structural Engineer, 1967, pp499-505
79(19), 2 October 2001, pp32-35
A.72 Short, W.D. ‘Accidents on construction work with
A.57 Edwards, T. et al. ‘Failure of roller bearings on the special reference to failures during erection or
Thelwall Viaduct’. In Neale, B.S. ed. Forensic demolition’. The Structural Engineer, 40(2), February
engineering: from failure to understanding. London: 1962, pp35-43
Thomas Telford, 2009, pp423-432
A.73 Almuñécar: seeds of a falsework failure’. New Civil
A.58 Health and Safety Executive. The track obstruction by a Engineer, 24 November 2005, p12
road vehicle and subsequent train collisions at Great
Heck 28 February 2001. Sudbury: HSE Books, 2002. A.74 Fennell, D. Investigation into the King’s Cross
Available at: http://www.rail-reg.gov.uk/upload/pdf/ underground fire. London: HMSO, 1988. Available at:
incident-greatheckfinal-optim.pdf [Accessed: http://www.railwaysarchive.co.uk/documents/DoT_
25 February 2013] KX1987.pdf [Accessed: 25 February 2013]

A.59 ‘Six hurt as tornado hits London’. BBC News website. A.75 Crawley, F. ed. Piper Alpha: lessons for life cycle safety
7 December 2006. Available at: http://news.bbc.co.uk/ management. Rugby: IChemE, 1990
1/hi/england/london/6217514.stm [Accessed:
25 February 2013] A.76 Lord Cullen. The Public Inquiry into the Piper Alpha
disaster. London: HMSO, 1990
A.60 Mr Justice Popplewell. Committee of inquiry into crowd
safety and control at sports grounds: final report. A.77 Home Office. The Hillsborough Stadium disaster, 15
London: HMSO, 1986 April 1989. Inquiry by Lord Justice Taylor: final report.
London: HMSO, 1990
A.61 Building Research Establishment. The Fire at Valley
Parade, Bradford [video]. Garston: BRE, 1987 A.78 Department for Culture, Media and Sport. Guide to
safety at sports grounds. 5th ed. [Norwich?]: The
A.62 Rose, D. ‘When the mortar doesn’t hold’. London Stationery Office, 2008
Review of Books, 22(6), 16 March 2000, p17.
Available at: http://www.lrb.co.uk/v22/n06/david-rose/ A.79 Explosion Mechanism Advisory Group report. London:
when-the-mortar-doesnt-hold [Accessed: 25 February Buncefield Major Incident Investigation Board, 2007.
2013] Available at: http://www.buncefieldinvestigation.gov.uk/
reports/buncefieldagr.pdf [Accessed: 25 February 2013]
A.63 ‘Body found after building collapse’. BBC News
website, 29 January, 2000. Available at: http://news. A.80 Department of Transport. Mv Free Enterprise: report of
bbc.co.uk/1/hi/uk/623342.stm [Accessed: 25 February Court no 8074. Formal investigation. London: HMSO,
2013] 1987. Available at: http://www.maib.gov.uk/
publications/investigation_reports/herald_of_free_
A.64 Legislative Council Panel on Planning, Land and enterprise.cfm [Accessed: 25 February 2013]
Works. Collapse of industrial building at 19 Sze Shan
Street, Yau Tong. Available at: http://www.legco.gov.hk/ A.81 Bureau d’Enquêtes et d’Analyses pour la Securité de
yr01-02/english/panels/plw/papers/209_02_-e.pdf l’Aviation Civile. Concorde accident: BEA accident
[Accessed: 25 February 2013] reports. Available at: http://www.concordesst.com/
accident/report.html [Accessed: 25 February 2013]
A.65 Holloway, S. Moorgate: anatomy of a railway disaster.
Newton Abbot: David & Charles, 1988 A.82 National Commission on Terrorist Attacks on the United
States. The 9/11 Commission Report. Available at:
A.66 Department of Transport. Report on the accident that http://govinfo.library.unt.edu/911/report/911Report.pdf
occurred on 28th February 1975 at Moorgate Station [Accessed: 25 February 2013]
on the Northern Line, London Transport Railways.
London: HMSO, 1976. Available at: http://www. A.83 Parker, D. ‘Control systems blamed for Severn gantry
railwaysarchive.co.uk/documents/DoE_Moorgate1975. failure’. New Civil Engineer, 2/9 June 1994, pp4-5
pdf [Accessed: 25 February 2013]
A.84 ‘Gantry fall deaths ‘unlawful’. BBC News website,
A.67 Lord Cullen. The Ladbroke Grove Rail Inquiry. Part 1 21 July 2003. Available at: http://news.bbc.co.uk/1/hi/
report. Sudbury: HSE Books, 2001. Available at: england/3084859.stm [Accessed: 25 February 2013]

62 The Institution of Structural Engineers Risk in structural engineering

 Case studies A.26

A.85 Institution of Structural Engineers. The Operation and

maintenance of bridge access gantries and runways.
2nd ed. London: IStructE, 2007

A.86 Nicholson, D. et al. The Observational method in

ground engineering: principles and applications. CIRIA
Report 185. London: CIRIA, 1999

A.87 Health and Safety Executive. The Collapse of NATM

Tunnels at Heathrow Airport. Sudbury: HSE Books,
2000

A.88 Standing Committee on Structural Safety. The Collapse

of NATM Tunnels at Heathrow Airport. SC/06/101.
Available at: http://www.structural-safety.org/
topicpapers [Accessed: 25 February 2013]

A.89 Report of the Presidential Commission on the Space

Shuttle Challenger Accident. Chapter V: The
Contributing cause of the accident. Available at: http://
history.nasa.gov/rogersrep/v1ch5.htm [Accessed:
25 February 2013]

A.90 Mason, R.O. ‘Lessons in organisational ethics from the

Columbia Disaster: can a culture be lethal?’.
Organizational Dynamics, 33(2), May 2004,
pp128-142

A.91 Columbia Accident Investigation Board. The Report.

Available at: http://www.nasa.gov/columbia/caib/html/
VOL1.html [Accessed: 25 February 2013]

A.92 Lew, H.S. et al. Investigation of construction failure of

Harbour Cay Condominium in Cocoa Beach, Florida.
NBS BSS 145. Washington, D.C.: NBS, 1982.
Available at: http://www.nist.gov/manuscript-
publication-search.cfm?pub_id=908924 [Accessed:
25 February 2013]

A.93 Canterbury Earthquakes Royal Commission. Final

report. Available at: http://canterbury.royalcommission.
govt.nz [Accessed: 25 February 2013]

A.94 ‘Fraudster’s Australian ties’. The Canberra Times, 22

October 2012. Available at: http://www.canberratimes.
com.au/national/fraudsters-australian-ties-20121021-
27zvy.html [Accessed: 25 February 2013]

A.95 U.S. Department of Energy, Office of Health, Safety

and Security. Suspect/Counterfeit Items Awareness
Training. Revision 6, June 2007. Available at:
http://www.hss.energy.gov/sesa/corporatesafety/sci/
SCIAwarenessTrainingManual062007.pdf [Accessed:
25 February 2013]

A.96 Kranakis, E. Constructing a bridge: an exploration of

engineering culture, design, and research in
nineteenth-century France and America. Cambridge,
Mass: MIT Press, 1997

A.97 Wood, J.G.M. ‘Paris airport terminal collapse: lessons

for the future’. The Structural Engineer, 83(5), 1 March
2005, pp13-14

The Institution of Structural Engineers Risk in structural engineering 63