Post-earthquake decision-making using influence diagrams

Michelle T. Bensi, Armen Der Kiureghian, Daniel Straub

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A Bayesian network (BN) is a probabilistic graphical model that represents a set of random variables and their probabilistic dependencies. When extended by decision and utility nodes into an influence diagram (ID), the resulting network can be highly useful for aiding post-earthquake decision-making. We develop preliminary IDs to solve a specific decision problem involving the post-earthquake inspection and shutdown of components. Specifically, we focus on a post-earthquake scenario in which an earthquake has occurred and placed seismic demands on the components of an infrastructure system and a decision-maker must decide, upon probabilistic updating for available information, whether to shutdown and/or inspect each component of the system. An example application demonstrates the methodology.

Original languageEnglish
Title of host publicationApplications of Statistics and Probability in Civil Engineering -Proceedings of the 11th International Conference on Applications of Statistics and Probability in Civil Engineering
PublisherTaylor and Francis Inc.
Pages343-351
Number of pages9
ISBN (Print)9780415669863
DOIs
StatePublished - 2011
Event11th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP - Zurich, Switzerland
Duration: 1 Aug 20114 Aug 2011

Publication series

NameApplications of Statistics and Probability in Civil Engineering -Proceedings of the 11th International Conference on Applications of Statistics and Probability in Civil Engineering

Conference

Conference11th International Conference on Applications of Statistics and Probability in Civil Engineering, ICASP
Country/TerritorySwitzerland
CityZurich
Period1/08/114/08/11

Fingerprint

Dive into the research topics of 'Post-earthquake decision-making using influence diagrams'. Together they form a unique fingerprint.

Cite this