Decision-support Measure for Resilient Infrastructure Networks Considering Operation Mechanism and Cascading Failure

Abstract

While the scale and frequency of disasters increase worldwide, infrastructure networks have become more complex as urbanization progresses. In efforts to maintain the networkﾒs performance through an appropriate decision-making process, disaster resilience is considered an essential ability of the systems. Recently, a system-reliability-based disaster resilience analysis framework was proposed to fully consider the relationship between the states of the system and the components. The proposed framework hinges on the reliability-redundancy (?-?) analysis, which enables a decision-maker to identify disruption scenarios critical to the system-level resilience. Based on the ?-? analysis method, this study aims to overcome limitations of existing component reliability importance measures in terms of adaptability to the target resilience of the system, considering the sequential component failures due to network flow redistributions, and excluding the spurious correlation caused by common source effects of the disastrous events. The proposed measure considers the effects of cascading failures by merging the causal effect of the state transition of the component in the disruption scenario on the change in the system state. Given the probabilistic information of the lines, electrical properties, and topology of the power grid, the proposed measure is computed to further support decision-making processes.