A Bayesian network framework for post-earthquake infrastructure system performance assessment

We are currently working toward the development of a probabilistic decision-support system (DSS) for near-real time emergency response and recovery following a seismic event. We utilize a Bayesian Network (BN) framework for assessing the performance of spatially distributed infrastructure systems subjected to seismic hazards. BNs have many valuable characteristics (e.g. they are graphical, adaptable, efficient, and enable Bayesian updating) that make them well-suited for such an application. We model seismic demands on an infrastructure system by constructing a BN model of ground motion intensity as a spatially distributed random field. In this paper we present this seismic demand model with recent enhancements including the addition of a finite source model. The performance of infrastructure system components and the overall system is modeled using fragility functions and BN system connectivity formulations. An example application to a hypothetical transportation system is presented to illustrate the power and value of the proposed BN framework.