Date of Award

5-2015

Degree Name

MS in Civil and Environmental Engineering

Department/Program

Civil and Environmental Engineering

Advisor

Robb Moss

Abstract

Natural hazards are a growing risk across the globe. As regions have urbanized, single events impact greater proportions of the population, and the populations within those regions have become more dependent on infrastructure systems. Regional resilience has become closely tied to the performance of infrastructure. For a comprehensive risk assessment losses caused by lifeline outage must be considered alongside structural and nonstructural risks. Many well developed techniques quantify structural and nonstructural risk; however, there are insufficient procedures to determine the likelihood of lifeline outages. Including lifelines in seismic assessments will provide a comprehensive risk, improving a decision maker’s capacity to efficiently balance mitigation against the full spectrum of risks.

An ideal lifeline risk assessment is infeasible due to the large geographic scale of lifeline systems and their system structure; these same characteristics also make them vulnerable to disruption in hazard events. Probabilistic methods provide solutions for their analysis, but many of the necessary analysis variables remain unknown. Continued research and increased collection of infrastructure data may improve the ability of advanced probabilistic methods to study and forecast performance of lifelines, but many inputs for a complete probabilistic model are likely to remain unknown. This thesis recognizes these barriers to assessment and proposes a methodology that uses consequences to simplify analysis of lifeline systems.

Risk is often defined as the product of probability of failure and consequence. Many assessments study the probability of failure and then consider the consequence. This thesis proposes the opposite, studying consequence first. In a theoretical model where all information is available the difference in approach is irrelevant; the results are the same regardless of order. In the real world however, studying consequence first provides an opportunity to simplify the system assessment. The proposed methodology starts with stakeholders defining consequences that constitute ruin, and then the lifeline system is examined and simplified to components that can produce such consequences. Previously large and expansive systems can be greatly simplified and made more approachable systems to study.

The simplified methodology does not result in a comprehensive risk assessment, rather it provides an abbreviated risk profile of catastrophic risk; risk that constitutes ruin. By providing an assessment of only catastrophic lifeline risk, the risk of greatest importance is measured, while smaller recoverable risk remains unknown. This methodology aligns itself with the principle of resilience, the ability to withstand shocks and rebound. Assessments can be used directly to consider mitigation options that directly address stakeholder resilience. Many of the same probabilistic issues remain, but by simplifying the process, abbreviated lifelines assessments are more feasible providing stakeholders with information to make decisions in an environment that currently is largely unknown.

Share

COinS