Developing a framework for the seismic risk assessment of interdependent lifelines systems for the insurance sector Indranil Kongar University College London UC Lifelines Week Day 2 21st April 2015
Single system assessment 1. Infrastructure system classification 2. Select damage scales 3. Determine appropriate hazard parameters 4. Generate seismic demand fields 5. Assign damage to system components 6. Measure system performance
2
Risk = Hazard x Vulnerability x Exposure HAZARD
Interdepende Independent nt
Earthquake Direct physical damage
Direct physical damage
VULNERABILITY
Element System AA
Element System BB Indirect operational outage
EXPOSURE
Direct impact: Repair cost
Indirect impact: Business interruption
Direct impact: Repair cost
Indirect impact: Business interruption
3
Catastrophe Models Loss = f(Exposure, Hazard, Vulnerability) Risk = f(Σ Loss x Probability of hazard)
4
Infrastructure taxonomy Infrastructure system Electric power
Potable water
Components
Component attributes
Generation plants Substations Cables Wells Water treatment plants Pumping stations Storage tanks
Capacity, seismic design level Voltage, seismic design level Material, size Seismic design level Capacity, seismic design level Capacity, seismic design level Elevation, material, geometry, seismic design level
Pipelines Lift stations Treatment plants Pipelines Natural gas Pipelines Compressor stations Fuel Refineries Pumping stations Storage tanks Pipelines Telecommunication Central offices Cables s Highways Roadways Bridges Waste water
Tunnels Embankments
Material, joint type, age, diameter Capacity, seismic design level Capacity, seismic design level Material, joint type, age, diameter Material, joint type, age, diameter Capacity, seismic design level Capacity, seismic design level Capacity, seismic design level Elevation, material, geometry, seismic design level Material, joint type, age, diameter Seismic design level Material, size Importance level Structural system, material, age, geometry, seismic design level Construction method, geometry, local geology Height, soil type
5
6
Interdependency • Two approaches to interdependency: a. Model connections between components within a systems, or: b. Model coupling between systems as a whole
System of systems
System coupling 7
Interdependency
Source: Reed et al. (2009)
Empirical Expert judgment
Source: Duenas-Osorio & Kwasinski (2012)
8
Case study: Christchurch Compare observed damage from September 2010 and February 2011 events with predictions from HAZUS
9
Buried cable fragility • No existing fragility functions for buried cables • Damage to cables observed from both September 2010 and February 2011 earthquakes so new functions can be derived
10
Observed liquefaction
11
Damage in non-liquefaction areas Typology
Ground shaking only
Minor liquefaction
Minor lateral spreading
Copper
0.08
0.23
0.44
Aluminium
0.02
0.02
0.32
PILCA
0.06
0.15
0.41
XLPE
0.01
0.00
0.20
Repair rates in zones where there was no observed ground deformation are lower than in areas where ground deformation was observed, indicating that ground deformation is a stronger driver of damage than ground shaking
12
Substations 1.0 0.9 0.8 0.7 0.6
0.4
New Min
0.3
HAZ Comp
0.2
New Comp
P (DS ≥ DSi)
0.5
HAZ Min
0.1 0.0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
Peak ground acceleration (g)
13
Thanks for your attention
[email protected] www.epicentreonline.co.uk
14