Systemic Validation of Consequence-Based Risk Management for Seismic Regional Losses

Steelman, Joshua; Hajjar, Jerome F.

Permalink

https://hdl.handle.net/2142/11998 Copy

Description

Title

Systemic Validation of Consequence-Based Risk Management for Seismic Regional Losses

Author(s)

• Steelman, Joshua
• Hajjar, Jerome F.

Issue Date

2008-09

Keyword(s)

• Mid-America Earthquake (MAE) Center
• Consequence-Based Risk Management (CRM)

Abstract

The Mid-America Earthquake (MAE) Center has performed much research since its inception relating to all facets of seismic loss assessment and risk management, from hazard definition through social and economic loss modeling. The culmination of this work is the integration of the research results into a comprehensive system for loss assessment, decision support, and consequence-based risk management (CRM). One vehicle through which the integration process takes form is MAEViz, a software program developed in a joint effort by the MAE Center and the National Center for Supercomputing Applications (NCSA). The purpose of this document is to present findings for systemic validation of the integration of research data threads within MAEViz. The validation effort is presented as part of the Memphis Testbed Project within the MAE Center, a comprehensive testbed in which much of the MAE Center research is being implemented. The general approach of the validation plan is to seek reports of risk assessments published in the literature that are sufficiently well documented that MAEViz can be used to perform a similar study. As part of the validation exercise, it was found that even relatively well documented studies rarely supplied sufficient data such that the study could be replicated in detail. The most suitable study was determined to be a risk assessment developed for the state of South Carolina using HAZUS, the program developed by the Federal Emergency Management Agency to perform loss assessments for natural disasters on the regional scale. The data reported in the South Carolina study has been used to define necessary parameters and execute a risk assessment for the region in MAEViz. Results of MAEViz analyses have been evaluated relative to the published results in this document, and it is found that while MAEViz is a suitable engine for performing risk assessments, the results of any risk assessment are sensitive to the specific algorithmic formulation implemented for the study. In this validation study, it was determined that differences in results obtained from MAEViz and HAZUS originate primarily from differences in the damage prediction and loss estimation methodologies. The damage estimation algorithm in MAEViz, using vulnerability formulations based on time history analyses of nonlinear structural response, provides an alternative estimation of damage prediction as compared to the Capacity Spectrum Method, a nonlinear static analysis method implemented in HAZUS, when determining probabilities of damage states. With regard to loss estimation algorithms, the general frameworks employed in MAEViz and HAZUS are similar, however, damage factors correlating damage states to economic losses are framed within a probabilistic context for MAEViz, leading to higher predicted damage for very lightly damaged structures, and lower predicted damage for very heavily damaged structures.

Type of Resource

text

Language

en

Permalink

http://hdl.handle.net/2142/11998

Sponsor(s)/Grant Number(s)

National Science Foundation EEC-9701785

Owning Collections