Presentation by H. Can Ünen @ ISCRAM 2009 Summerschool

1. Seismic Performance
Assessment of
Interdependent Lifeline
Utility Systems
Hüseyin Can Ünen
Istanbul Technical University, Turkey

2. Born in 1982, Ankara, Turkey.
B.Sc.: Civil Engineering, Middle East Technical
University, Ankara (2004).
M.Sc.: Satellite Communications and Remote
Sensing, Istanbul Technical University (2006).
Currently a Ph.D. Student in Geomatics Engineering,
Istanbul Technical University.
Study topics: Surveying, Geographic Information
Systems, Disaster Management.

3. Why Lifeline Networks?
Electric power, potable water,
communication, transportation, natural
gas, waste water, etc.
Vital to the health, safety, and social
activities of the community. Also vital to
the functioning of an urban industrialized
society.
Serviceability of power, water and
communication systems are essential for
survival and also for response and
recovery efforts following a disaster.

4. INTERDEPENDENCY
Interdependency: A bidirectional relationship between
two infrastructures through which the state of each
infrastructure influences or is correlated to the state of the
other.
or
The connections among agents in different infrastructures
in a general system of systems.

5. INTERDEPENDENCY

6. Aim: To achieve more accurate and reliable
seismic performance assessment of lifeline
utility networks.
Water network Power network

7. Interdependent Network Analysis (INA) Model
Topological Model Structural Model
INVENTORY
HAZARD FRAGILITY
SYSTEM
CONNECTIVITY
STRUCTURAL
DAMAGE
Monte Carlo Simulations (n)
COMPONENT
FAILURE
RE-STRUCTURING
OF NETWORK
PERFORMANCE Connectivity Loss
ASSESSMENT
Service Flow Reduction

8. Determination of Structural Damage
Estimation of damage levels to the structures are made by
implementing fragility functions, which give the probability
that a limit state is exceeded, or by damage functions giving
the amount of expected damage, given an input level of
shaking.

9. Structural Damage Analysis
Damage levels to the structures are estimated by using
given level of ground shaking and
Fragility functions: the probability that a limit state is exceeded,
OR
Damage functions: the amount of expected damage
Results: Hazard
Definition
Inventory
Selection
Fragility
Models
Damage state probability
Expected damage
Damage Analysis
Repair rate
Break rate
Leak rate Structural Damage

10. Topologically modeled networks are
built of links (pipelines, power lines)
Water network facilities Water pipelines
and nodes (network facilities).
Modeling can be done if the
connectivity and flow patterns of the
system are known.
System performances are assessed
by applying connectivity and flow
algorithms.
Water network

11. System Performance Measures
CONNECTIVITY LOSS
Quantifies the decrease in the number of generation
facilities with connecting paths to the distribution facilities.
Calculation of the parameter relies on the topological
structure of the network and the existence of paths
connecting supply and demand elements.

12. System Performance Measures
SERVICE FLOW REDUCTION
Quantifies the amount of flow that does not meet the
distribution vertex demands.
Adresses the impact on the end users.
Demands and capacities must also be known in addition to
the network topology.

13. Interdependent Network Analysis (INA) Model
Topological Model Structural Model
INVENTORY
HAZARD FRAGILITY
SYSTEM
CONNECTIVITY
STRUCTURAL
DAMAGE
Monte Carlo Simulations (n)
COMPONENT
FAILURE
RE-STRUCTURING
OF NETWORK
PERFORMANCE Connectivity Loss
ASSESSMENT
Service Flow Reduction

14. System Outcomes
When the expected damage to pipelines,
facilities, and buildings known, they can be
used for:
Improving resiliency of the networks.
Developing retrofit strategies.
Estimating the repair cost and workmanship
needed.

15. THANK YOU