Bayesian approach for the network reconstruction of interdependent critical infrastructure systems from cascading failures

Analyzing the behavior of complex interdependent networks requires complete information about the network topology and the interdependent links across networks. For many applications, such as interdependent critical infrastructure (ICI) networks, understanding network interdependencies is crucial to anticipate cascading failures and mitigate the risk from disruptions. However, complete network data are often unavailable due to security concerns, and some important interdependent links are revealed only in the aftermath of a disruption. This study formulates and solves a network reconstruction problem to uncover uncertain network interdependencies during disruptions. We propose a scalable nonparametric Bayesian approach to reconstruct the topology of ICI networks from (observed) cascading failures. Metropolis-Hastings (M-H) algorithm coupled with the infrastructure-dependent proposal is employed to increase the efficiency of sampling possible graphs. Numerical results of reconstructing a synthetic system of ICI networks demonstrate that the proposed approach outperforms existing methods in both accuracy and computational time.