Model and solution method for mean-risk cost-based post-disruption restoration of interdependent critical infrastructure networks

Critical infrastructure networks (CINs), such as power grids, water distribution systems, and telecommunication networks, are essential for the functioning of society and the economy. As these infrastructure networks are not isolated from each other, their functions are not independent and may be vulnerable to disruptive events (e.g., component failures, terrorist attacks, natural disasters). For decision makers, how to restore the functions of CINs while accounting for interdependencies and various uncertainties becomes a challenging task. In this work, we study the post-disruption restoration problem for a system of interdependent CINs under uncertainty. We propose a two-stage mean-risk stochastic restoration model using mixed-integer linear programming (MILP) with the goal of minimizing the total cost associated with unsatisfied demands, repair tasks, and flow of interdependent infrastructure networks. The restoration model considers the availability of limited time and resources and provides a prioritized list of components to be restored along with assigning and scheduling them to the available network-specific work crews. Additionally, the model features flexible restoration strategies including multicrew assignment for a single component and a multimodal repair setting along with the consideration of full and partial functioning and dependencies between the multi-network components. The proposed model is illustrated using the power and water networks in Shelby County, Tennessee, United States, under two hypothetical earthquake scenarios.