Comparing decision models for disaster restoration of interdependent infrastructures under uncertainty

Abstract

As infrastructure systems are highly interdependent, one needs to analyze their disaster resilience and develop restoration plans with the consideration of infrastructure interdependencies. This study presents two probabilistic models for infrastructure decision-makers to simulate the recovery of interdependent systems in a post-disaster scenario. The models consider interdependencies related to damage, functionality, and restoration. To incorporate uncertainty in restoration, this study assumes that the actual duration of each restoration activity follows a random distribution. To simulate the decision-making process in the recovery, this study uses the PRAISys platform to implement two schemes with different restoration criteria. The first scheme uses the priority ranking of the damaged structure (based on its importance, criticality, etc.) as the criterion and the platform simulates the restoration plan under resource and dependency constraints. In contrast, the second uses the criterion of minimizing the restoration completion time by framing all restoration activities within a constrained optimization formulation, and the platform implements the optimal schedule as the restoration plan. To exemplify the applicability of the two schemes, this study simulates the recovery of interdependent systems after a hypothetical earthquake in the Lehigh Valley, a multi-county community in eastern Pennsylvania, USA.