Integrated Bayesian Network and Strongest Path Method (BN-SPM) for effective multi-hazard risk assessment of interconnected infrastructure systems

Abstract

This study proposes an integrated framework based on Bayesian Network (BN) and Strongest Path Method (SPM) to provide a robust assessment of the natural hazard risks to interconnected infrastructure systems. Leveraging Bayesian inference, the BN transforms prior failure probabilities into posterior ones, while SPM prioritizes network entities based on global impact, global vulnerability, and risk indices, calculated from key parameters including the likelihood of failure, degree of impact, and degree of dependence. The proposed integrated BN-SPM model calculates risk indices for complex interconnected networks with conditional dependencies on external nodes, such as multi-hazard, or mitigation and redundancy strategies, without predetermined parameters. Using this framework we evaluate diverse scenarios in Saint Lucia, encompassing non-hazard condition, flooding hazards, and presumptive strategies like mitigation and redundancy. The high likelihood of failure and complex interdependencies heighten risks even under non-hazard condition. Particularly, the Hewanorra International Airport (HIA) emerges as the most vulnerable entity. The established risk assessment framework offers the potential for application in evaluating risks within complex interconnected systems such as power grids, transportation networks, and communication systems facing various multi-hazard scenarios. This framework also assesses the implementation of mitigation and redundancy strategies, reducing the risk index within the affected network domain.