Dr Sarah Dunn, Lecturer in Structural Engineering, Newcastle University, UK visited SMART Infrastructure Facility on Friday, August 21st 2015. During her visit, Dr Dunn presented a summary of her research as part of the SMART Seminar Series.

1. How Resilient are our Infrastructure Systems?
Dr Sarah Dunn
How do we Increase their Resilience?
Lecturer in Structural Engineering

2. Introduction

3. Research Team (Selected)
David Alderson
Research Associate in GeoInformatics
Dr Stuart Barr
Senior Lecturer in GIS
Prof Hayley Fowler
Professor in Climate Change Impacts
Dr Sean Wilkinson
Senior Lecturer in Structural Engineering
Prof Richard Dawson
Professor of Earth Systems Engineering
Dr Jaime Amezaga
Reader in Environmental Sustainability
Ali Ford
Researcher in Geomatics
Dr Gaihua Fu
Research Associate

4. Introduction
• Sept 2006 – July 2010
MEng Civil and Structural Engineering
• Sept 2010 – Feb 2014
PhD in Civil Engineering
‘An Investigation to Improve Community Resilience using
Network Graph Analysis of Infrastructure Systems’
• March 2014 – March 2015
EPSRC Doctoral Prize Research Fellow
‘Increasing Community Resilience to Climate Change
Impacts through Adaptation of Infrastructure Systems’
• March 2015 – Present
Lecturer Structural Engineering

5. Main Research Area
Aim of Research
To improve the resilience of our communities by developing techniques that can identify
fragile system architectures, recognize vulnerable areas within these systems and
establish methods that can help to protect them from hazard.

6. Engineering Grand Challenges

7. Fragility of Infrastructure Systems
North American Blackout, 2003
• Affected 50 million people, in 8 US states
• Up to 4 days to restore power
• Economic losses between $7 and $10 billion US Dollars
http://earthobservatory.nasa.gov/images/imagerecords/3000/3719/NE_US_OLS2003227.jpg

8. Fragility of Infrastructure Systems
http://www.huffingtonpost.com/2013/08/14/2003-northeast-blackout_n_3751171.html

9. Analysis of Infrastructure Systems
Underlying
Network
Architecture
Vejvodova, J. (2006). Drinking Water [online]. [Accessed 28/11/09].

10. Scale-Free Networks
• The Internet
(Albert, et al. Nature: 1999)
• The World-Wide-Web
(Albert, et al. Nature: 2000)
Newman (2003)
• Electrical Distribution Systems
(Sole, et al. 2008)
• Airline Networks
(Wilkinson, Dunn and Ma 2012)
Carvalho, et al. (2009)
Exponential Networks
Albert, R., Jeong, H. and Barabasi, A. L. (2000). "Error and Attack Tolerance of Complex Networks."
Nature 406(6794): 378-382.

11. Previous Application of Network Theory
1. Communication Networks (Albert et al. 1999; Albert et al. 2000; Cohen et al. 2000)
2. Power Grids (Crucitti et al. 2004; Holmgren 2006; Sole et al. 2008)
3. Transport Networks (Guimera and Amaral 2004; Mathe et al. 2013; Tu et al. 2013)
4. Water Distribution Systems (Walski 1993; Yazdani and Jeffrey 2011)
However, the spatial component of these real world networks
receives less attention… when not neglected entirely.

12. Case Study: Eyjafjallajökull Volcano
20th March 2010 Eruption
• Disrupted European air space
between 14th and 21st April
• Caused 10 million air
passengers to be delayed
• Estimated economic losses
1.7 billion US Dollars

13. Disruption to European Air Travel
18th April
21st April
525 Airports
3886 Air routes

14. European Air Traffic Network

15. European Air Traffic Network

16. Network Generation Algorithm
Scale-free network generation algorithm:
Barabasi, A. L. and Albert, R. (1999). "Emergence of scaling in random networks." Science 286(5439): 509-512.

17. Network Generation Algorithm
Exponential network generation algorithm:
Wilkinson, S. M., Dunn, S., and Ma, S., (2012) ‘The Vulnerability of the European Air Traffic Network to Spatial
Hazards’ Natural Hazards. 60(3): 1027-1036.

18. European Air Traffic Network

19. Spatial Characteristics of EATN

20. Hazard Tolerance of Synthetic Networks

21. Hazard Tolerance of Synthetic Networks

22. Specific Vulnerable Components

23. Specific Vulnerable Components
𝐶𝑀2 =
𝑂𝑟𝑖𝑔𝑖𝑛𝑎𝑙 𝐹𝑙𝑜𝑤 ∗ 𝐵𝑒𝑡𝑤𝑒𝑒𝑛𝑛𝑒𝑠𝑠 𝐶𝑒𝑛𝑡𝑟𝑎𝑙𝑖𝑡𝑦
𝐷𝑒𝑔𝑟𝑒𝑒
D : Degree of Node
OF : Original Flow
BC : Betweenness Centrality
+ 2 combinations of these metrics
0.0
0.2
0.4
0.6
0.8
1.0
0 5 10 15
R2ValueforMeasures
Supply Node
D
OF
BC
CM1
CM2
Dunn, S., and Wilkinson, S. M., (2012)
‘Identifying Critical Components in Infrastructure
Networks using Network Topology’ ASCE Journal
of Infrastructure Systems.19(2): 157-165.

24. Other research…
Hazard tolerance of other air traffic networks
• Also analyzed the hazard tolerance of the China and US air traffic
networks.
• Dunn, S., Wilkinson, S. M., and Zhang, H., (2013) ‘Methods to Assess the Hazard Tolerance of Spatial Networks’
International Conference on Civil, Environmental and Infrastructure Engineering, 2013. Copenhagen, Denmark.
Development of methods to increase the resilience of air traffic networks
• We developed an ‘adaptive’ and a ‘permanent’ resilience strategy.
• Finding that the ‘adaptive’ strategy showed the most potential.
• Dunn, S., and Wilkinson, S. M., (2015) ‘Increasing the Resilience of Air Traffic Networks using a Network Theory
Approach’. Transportation Research Part B (under review)
Analysis of Interdependent Infrastructure Systems
• Considered new methodology to assess the impact of failure in one
system upon a dependent system.
• Dunn, S., Holmes, M., and Wilkinson, S. M., (2014) ‘Modelling Interdependent Cascading Failures in Real World
Complex Networks using a Functional Dependency Model’ International Symposium for Next Generation
Infrastructure 2014. Vienna, Austria.

25. Network Graph Theory
 Very good at providing a level of confidence that the system will perform well to
untested / unanticipated scenarios
X Does not model ‘flow’ of resources and therefore it can be difficult to quantify
the loss of supply to consumers, or to calculate economic impact with any
certainty
This can be achieved using a Catastrophe Risk Modelling framework…

26. CAT Modelling
• Weather event: wind storm, heatwave, rainfall
• Information: duration, intensity, location
Event
Generation

27. CAT Modelling
Event
Generation
Intensity
Calculation
• Calculation of event intensity (e.g. wind
speed, temperature) at each individual
component location
Showing the maximum wind speeds during the Burns Day Storm on the 24th January 1990, shown on a red-green scale,
where red indicates areas of high wind speed and green areas of low wind speed.

28. CAT Modelling
Event
Generation
Intensity
Calculation
Exposure
Information
• Data regarding infrastructure
component type and location

29. CAT Modelling
Event
Generation
Intensity
Calculation
Exposure
Information
Damage
Estimation

30. CAT Modelling
Event
Generation
Intensity
Calculation
Exposure
Information
Damage
Estimation
Consequence
Calculation
0
5
10
15
20
25
100 200 300 400 500 600 700 800 900 1000
Liklihood
Estimated Economic Cost (£)

31. Fragility / Vulnerability Curves
0
1
2
3
4
5
-4 1 6
P(f)
Flood Depth (m)
Damage
Estimation
• These curves define the relationship between the magnitude
of an event (e.g. maximum wind speed or flood depth) and
the probability of failure for individual components.
0
1
2
3
4
5
-4 1 6
P(f)
Wind speed (m/s)

32. UK Electrical Distribution Infrastructure

33. UK Electrical Distribution Infrastructure
Exposure
Information

34. UK Electrical Distribution Infrastructure
Towers / Poles
Substations
Overhead Lines
Underground Cables
Exposure
Information

35. Fragility / Vulnerability Curves
• We have worked with one DNO in the
UK to develop fragility curves for their
overhead line components
• These were developed using empirical
fault data, held in the NaFIRS database
• This records data regarding the date,
cause of fault, duration, approximate
fault location and number of consumers
affected
• However, it does not record the wind
speed which caused the fault

36. Wind Data
• In the UK the Met Office records data
regarding wind speed gusts (averaged
over a 3 second interval) at various
locations
• However, not all of these records are
simultaneous or continuous and there
are often large periods of time with no
data
• There are also areas of the UK that are
not ‘covered’ by an observation station
Image plotted using data from the Met Office

37. Fragility / Vulnerability Curves
• Therefore we have used wind velocities
generated using KNMI data
• Koninklijk Nederlands Meteorologisch
Instituut, Dutch: Royal Netherlands
Meteorological Institute
• These form a regular grid (~12km) of
observations over the UK
• For further detail:
• Wilkinson, S. M., Fowler, H., Manning, L., and
Dunn, S. (2014) "ECLISE Deliverable 4.07 of
Task T4.5: Estimates of Extreme Wind Speed
used for the Design of Buildings."

38. Identification of Wind Storms
0
2
4
6
8
10
12
0
5
10
15
20
25
30
0 50 100 150 200 250 300
NumberofFaults
WindSpeed(m/s)
Time Series (hours)
Wind Speed
Threshold Value
Wind Storm 1
Wind Storm 2
Number of Faults
• We identify individual wind storms, based on a threshold value of wind speed
(obtained from literature).
• We then attribute all faults to the maximum wind speed recorded during the
storm.

39. Identification of Wind Storms
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0 10 20 30 40
AverageNumberofFaultsperLengthof
OHLx10-3(inkm)
Wind Speed (m/s)
R² = 0.9889
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0 10 20 30 40
AverageNumberofFaultsperLengthof
OHLx10-3(inkm)
Wind Speed (m/s)

40. Fragility / Vulnerability Curves
R² = 0.9889
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0 10 20 30 40
AverageNumberofFaultsperLengthof
OHLx10-3(inkm)
Wind Speed (m/s)
• We therefore obtain the following vulnerability curve for overhead lines:

41. Fragility / Vulnerability Curves
R² = 0.9889
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
0 10 20 30 40
AverageNumberofFaultsperLengthof
OHLx10-3(inkm)
Wind Speed (m/s)
0
5
10
15
20
25
30
020406080
Frequency
NumberofFaults

42. Fragility / Vulnerability Curves
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0 10 20 30 40
CoefficientofVariation
Wind Speed (m/s)
0.000
0.002
0.004
0.006
0.008
0.010
0.012
0.014
0.016
0.018
0.020
0 10 20 30 40
StandardDeviation(x10-3) Wind Speed (m/s)

43. Fragility / Vulnerability Curves
0
20
40
60
80
100
120
140
160
180
17 19 21 23 25 27 29 31 33 35 37
NumberofStorms
Wind Speed (m/s)
Without Faults
With Faults
• We can also consider the proportion of storms where faults occurred:

44. Fragility / Vulnerability Curves
R² = 0.9889
R² = 0.9833
R² = 0.9846
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0 10 20 30 40
AverageNumberofFaultsperLengthof
OHLx10-3(inkm)
Wind Speed (m/s)
Method 2
Method 3: Urban
Method 3: Rural
• We considered ‘splitting’ the vulnerability curve for different land usage types:

45. Summary

46. Thank you.