A hierarchical Bayesian approach for assessing infrastructure networks serviceability under uncertainty: A case study of water distribution systems

Measuring the performance of infrastructure networks is critical to allocation resources before, during, and after a system’s disruption. However, lack data often hinders ability accurately estimate performance, resulting in uncertainty its evaluation which can lead biased estimates. To address this challenge, study develops Bayesian approach measure network at component level incorporate it system-level serviceability. Component fragility metrics are estimated using hierarchical model then integrated into system serviceability assessment Monte Carlo simulation shortest-path algorithm. These measures be dynamically updated as more becomes available. A case water distribution Shelby County Tennessee subject earthquake flood hazards presented illustrate proposed approach. Results show that topology important determining functionality under seismic hazard while vulnerability dominant factor hazard.