A presentation conducted by Matthew Holmes, Civil Engineering & Geosciences Newcastle University. Presented on Tuesday the 1st of October 2013. As infrastructure becomes increasing integral to daily lives, society becomes more vulnerable to potential failures. We mitigate against this by investing some of the increased prosperity afforded by infrastructure to treat the most salient risks and increase the resilience of the system. Therefore we enter a cycle whereour ability to identify and prioritise vulnerabilities is crucial to the future development of infrastructure. It is easy to compose a list of risks occupying the whole spectrum from probable through to fanciful, but, how do infrastructure owners define defensible boundaries between the credible risks they should assess and those that can be set aside? This paper tests the hypothesis that incorporating information on the uncertainty of risk assessments provides risk managers with a more robust process to justify their choice of credible risks.

1. ENDORSING PARTNERS
How do we ensure the
assessment of
infrastructure resilience is
www.isngi.org
proportionate to the risk?
The following are confirmed contributors to the business and policy dialogue in Sydney:
•
Rick Sawers (National Australia Bank)
•
Nick Greiner (Chairman (Infrastructure NSW)
Monday, 30th September 2013: Business & policy Dialogue
Tuesday 1 October to Thursday, 3rd October: Academic and Policy
Dialogue
Presented by: Matthew Holmes, Civil Engineering & Geosciences
Newcastle University
www.isngi.org

2. How do we ensure the assessment of
infrastructure resilience is proportionate
to the risk?
Matthew Holmes a, Anna Provost b, Derek Clucas b, Sean Wilkinson a
a. Civil Engineering & Geosciences, Newcastle University, Newcastle-upon-Tyne, NE1 7RU
b. United Utilities Plc., Lingley Mere Business Park, Warrington, WA5 3LP
International Symposium for Next Generation Infrastructure
Wollongong, Australia
1st October 2013
2

3. Industrial Doctorate - Acknowledgements
Dr. Sean Wilkinson
Prof. Richard Dawson
David Owen
Dr. Matt Hill
Anna Provost
Derek Clucas
Mark Jones
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
3

4. The challenge facing
infrastructure risk managers
4

5. The Risk Manager’s Challenge:
Development & Infrastructure
More demand
More capital to invest
Development
Infrastructure
provision
Infrastructure
supports development
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
5

6. The Risk Manager’s Challenge:
Vulnerability & Resilience
Greater
dependency on
infrastructure
Society is more
vulnerable to
failure
Development permits investment
Development
Infrastructure
provision
More reliable
infrastructure
Infrastructure
needs to be
more resilient
Investment
treats the most
salient risks
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
6

7. The Risk Manager’s Challenge:
Risk Identification
Risk managers have
to consider an
increasing wide
range of risks
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
7

8. The Risk Manager’s Challenge:
Risk Selection
Photo: Kyodo/Reuters
Photo: AP
Managers have to
consider an
increasingly wide
range of risks
They cannot
assess every
risk in detail
Their selection
process must
be robust and
accountable
Photo: Daily Telegraph
Photo: AP
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
8

9. Alternative approaches to
selecting risks
1. Likelihood
vs.
Consequence
2. Subjective
appraisal of
uncertainty
3. Formalised
appraisal of
uncertainty
9

10. Likelihood vs. Consequence
10

11. Approaches to selecting risks:
Likelihood versus Consequence
High consequence
risks facing the
United Kingdom
(after UK Cabinet Office
2010, red line added)
Risks of natural hazards
and major accidents
(UK Cabinet Office 2012)
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
11

12. Approaches to selecting risks:
L vs. C, with uncertainty
Indicative National
Risks
N.B. Axes reversed
compared to the UK version
(IPENZ 2012)
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
12

13. Subjective Appraisal of
Uncertainty
13

14. Cost of
obtaining
information
Value of
information
obtained
The problem with highly uncertain
appraisals
Uncertainty
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
14

15. Approaches to selecting risks:
Subjective Appraisal of Uncertainty
Risk selection
based upon the
uncertainty of the
probability
assessment
(adapted from the UK
Cabinet Office diagram in
slide 9)
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
15

16. Formal Appraisal of Uncertainty
16

17. Approaches to selecting risks:
Formalised Appraisal of Uncertainty
Loss of access to
PS1
P(Snow & ice)
Snow & ice
Loss of access to
PS3
Loss of access to
WwTW
P(Cold weather)
Cold weather
Loss of power to
PS1
Storms & gales
Cyber attacks:
infrastructure
Public disorder
Loss of access to
PS2
P(Loss of power)
Loss of power
Loss of power to
PS2
Loss of power to
PS3
Loss of power to
WwTW
P(Loss of access
to PS1)
P(Loss of access
to PS2)
P(Loss of access
to PS3)
P(Loss of access
to WwTW)
P(Loss of power
to PS1)
P(Loss of power
to PS2
P(Loss of power
to PS3)
P(Loss of power
to WwTW)
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
17

18. Approaches to selecting risks:
Results
Cold
Weather
Storms &
Gales
• High risk
• Large spread of
results: low
precision
• Medium – high
risk
• Probability
restricted to a
narrow band
Cyber Attacks:
Infrastructure
Public
Disorder
• Low risk
• Large
uncertainty
• Very low risk
• Low uncertainty
(only 1 failure
mode)
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
18

19. Approaches to selecting risks:
Comparison with Likelihood vs. Consequence
Likelihood versus Consequence
2.0
Cold
Weather
1.5
1.0
Public
0.5
Disorder
0.0
0.00
Cyber Attack:
Infrastructure
Storms &
Gales
2.0
Cold
Weather
1.5
1.0
Cyber Attack:
Infrastructure
0.5
Public
Disorder
0.0
0.02
0.04
0.06
0.08
Mean:
Probability of at least one failure
•
•
•
2.5
Storms &
Gales
Mean: Flooding Volume [ML]
Mean: Flooding Volume [ML]
2.5
Uncertainty versus Consequence
0.10
1.0
0.8
0.6
0.4
0.2
0.0
Coefficient of Variation:
Probability of at least one failure
Does the new information materially affect the selection?
Is this extra information worth the extra effort?
How reliable is a risk manager’s judgement?
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
19

20. Discussion & Conclusion
20

21. Discussion
1. Omitting uncertainty at the risk screening phase is not
inherently flawed.
2. However, it misses an opportunity to justifiably eliminate some
risks whose assessment adds little value.
3. Incorporating subjective information on the uncertainty over
probability is more perceptive, but subjective and opaque.
4. Formalising this process reduces opacity, but the extra
complexity and cost may exceed the value of the new
information.
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
21

22. Conclusion
1.
The cost of assessing
marginal risks
We are more reliant on
infrastructure
We are more vulnerable
to failure
We need to assess
increasingly marginal
risks
We require new
methods required to
select risks to assess
2.
The more formal appraisal of uncertainty is interesting but flawed
because:
a) It does not eliminate subjectivity, just moves and exposes it.
b) The effort required to do it is self-defeating, given we are
looking for new ways to screen risks.
3.
Therefore, the subjective appraisal looks the most promising
approach.
Civil Engineering and Geosciences • Cassie Building • Newcastle University • Newcastle upon Tyne • NE1 7RU • UK • www.ceg.ncl.ac.uk
For further information; m.j.holmes@ncl.ac.uk
22

23. Thank you
References:
Any questions?
Cabinet Office (2013) National risk register of civil
emergencies 2013 edition. London: Cabinet Office.
matthew.holmes@stream-idc.net
Cabinet Office (2011) Keeping the country running: natural
hazards and infrastructure. London: Cabinet Office.
IPNEZ (2012) A safer New Zealand: reducing our exposure
to natural hazards. Wellington: IPENZ
23