ABSTRACT

Infrastructure systems are critical for a functioning society. They are also becoming increasingly connected, with the functioning of one system dependent on the states of the components in others. This chapter presents a probabilistic framework for modeling interdependencies for critical infrastructure systems to assess and improve resilience. First, three general, comprehensive types of interdependencies are defined. These capture the functional, geographic, and recovery dependencies between multiple infrastructure systems serving a given community. Each of these is then modeled using a probabilistic Bayesian network framework. This enables the dependencies between components both within a system and across systems to be quantitatively modeled. It also enables rapid updating of system assessments with new information. The framework is applied to an example interdependent water, power, and gas network to demonstrate its use. Inference results are provided to show the types of analyses possible using the proposed method to assess and improve the resilience of infrastructure systems.