Resilience-driven post-disaster restoration of interdependent infrastructure systems under different decision-making environments

Abstract

Critical infrastructure systems are highly interconnected and mutually dependent for smooth functioning. Such interdependencies contribute to operational efficiency but may also exacerbate the negative impacts caused by disruptions, as the failure of one system could spread to its connected systems. To enhance the resilience of interdependent infrastructure systems, this article investigates the post-disaster restoration decision problem and considers two decision-making environments. Firstly, a deterministic restoration decision model is developed under certainty to seek a combined repair sequence that can maximize the resilience of the interdependent system. This model assumes that the decision-makers have perfect information about the restoration decision problem. Then, this article extends this deterministic model to a two-stage stochastic restoration model under uncertainty, in which the repair time of damaged components is assumed to be random and represented by a set of scenarios. A heuristic method, composed of a selection principle and a matrix-based approach, is proposed to solve these two restoration decision models. Numerical experiments on interdependent systems demonstrate that integrating interdependency into the restoration decision problem could significantly benefit system resilience. The developed restoration decision models and heuristic method could provide essential insights into the restoration process of interdependent infrastructure systems.