Matrix-based System Reliability Method

Risk assessment of urban infrastructure networks under natural and man-made hazards is often performed by repeated computational simulations based on random samples of hazard intensities and corresponding component status. This sampling-based approach may prevent near-real-time application of probabilistic methods to hazard mitigation of infrastructure networks and risk-informed decision support. This paper proposes a non-sampling-based, multi-scale network reliability analysis approach based on the recently proposed matrix-based system reliability method. This approach accounts for the spatial correlation of hazard intensities and makes use of the disjoint cut sets or link sets identified by a state-of-the-art network analysis algorithm. The paper demonstrates the proposed approach through a case study of a Memphis Light, Gas and Water natural gas network under seismic hazard.