Bayesian Framework for Performance Assessment and Risk Management of Transportation Systems subject to Earthquakes

Project # NCTRAD

Research Team

  • Armen Der Kiureghian, Professor, UC Berkeley (PI)
  • Michelle Bensi, Graduate Student Researcher (GSR), UC Berkeley

Research Abstract

A Bayesian Network (BN) methodology is developed for performing infrastructure seismic risk assessment and providing decision support with an emphasis on immediate post-earthquake ap-plications. A BN is a probabilistic graphical model that represents a set of random variables and their probabilistic dependencies. The variables may represent demand or capacity values, or the states of components and systems. Decision and utility nodes may be added that encode various decision alternatives and associated costs, thus facilitating support for decision-making under uncertainty.

BNs have many capabilities that make them well suited for the proposed application. Most important among these is the ability to update the probabilistic states of the variables upon receiving relevant information. Evidence on one or more variables, e.g., measured component capacities or demands, or observed states of components, can be entered into the BN and this information propagates throughout the network to provide up-to-date probabilistic characterizations of the infrastructure components and system as well as optimal ordering of the decision alternatives. This can be done in near-real time and under the uncertain and evolving state of information that is characteristic of the post-event period. As is the case with most computational methods, BNs have their limitations: calculations can be highly demanding when the BN is densely connected, or when the infrastructure system is complex and large. This study addresses these challenges.

The proposed methodology consists of four major components: (1) a seismic demand model of ground motion intensity as a spatially distributed random field, accounting for multiple sources and including finite fault rupture and directivity effects, (2) a model for seismic performance of point-site and distributed components, (3) models of system performance as a function of component states, and (4) models of post-earthquake decision-making for inspection and operation or shutdown of components.

Two example applications demonstrate the proposed BN methodology. The second of these em-ploys a hypothetical model of the proposed California high-speed rail system subjected to an earthquake.

Research Outcomes

PEER reports:

Archival Journal Papers:

  • Bensi, M., A. Der Kiureghian and D. Straub (2011). Bayesian network modeling of correlated random variables drawn from a Gaussian random field. Structural Safety, to appear.

Papers in Conference Proceedings:

  • Bensi, M., A. Der Kiureghian and D. Straub (2009). A Bayesian network framework for post-earthquake infrastructure system performance assessment. Proceedings, ASCE Technical Council on Lifeline Earthquake Engineering Conference, June 28-July 1, 2009, Oakland, CA, page 1096 (CD-ROM)
  • Bensi, M., D. Straub, P. Friis-Hansen and A. Der Kiureghian (2009). Modeling infrastructure system performance using BN. Proc. 10th Int. Conf. On Structural Safety and Reliability, Osaka, Japan, September 13-17, 2009, CRC Press, Boca Raton, FL (CD-ROM, pp. 2843-2850).
  • Der Kiureghian, A., M. Bensi and D. Straub (2009). Bayesian network methodology for post-earthquake infrastructure risk management. Chapter 9 in Frontier Technologies for Infrastructures Engineering, S-S. Chen and A. H-S. Ang Editors., CRC Press, Taylor & Francis Group, London, U.K.
  • Bensi, M., A. Der Kiureghian and D. Straub (2010). Bayesian network for infrastructure seismic risk assessment: challenges and opportunities. Proceedings, 7th International Conference on Urban Earthquake Engineering & 5th International Conference on Earthquake Engineering, Tokyo, Japan, March 3-5, 2010.
  • Bensi, M., A. Der Kiureghian and D. Straub (2010). Bayesian network modeling of system performance. Proceedings,15th IFIP WG 7.5 Working Conference on Reliability and Optimization of Structural Systems, D. Straub Ed., Technical University of Munich, Germany, April 7-10, 2010, pp. 1-8.
  • Bensi, M., and A. Der Kiureghian (2010). Seismic hazard modeling by Bayesian network and application to a high-speed rail system. Proceedings of International Symposium on Reliability Engineering and Risk Management, J. Li et al. Edts., Tongji University Press, Shanghai, China, September 23-26, 2010.

Research Impact

An entirely new approach for seismic risk assessment and management of infrastructure systems has been developed. Employing the methodologies of Bayesian network and decision graphs, the approach offers a versatile and effective means for modeling infrastructure systems and assessing and managing seismic risk. It provides an efficient framework for updating probabilistic charac-terization of components and systems in light of evolving and uncertain information gained from measurements (e.g., recorded ground motions) and observations (e.g., observed states of compo-nents and system). Furthermore, it provides a means for prioritizing decision alternatives in light of updated state probabilities, given utility values associated with different outcomes and deci-sion alternatives. The framework is ideal for application to post-earthquake risk assessment and decision-making for damage mitigation and restoration of services.