Computational Framework for Time-Dependent Reliability Assessment of Underground Water Distribution Pipeline Networks

This paper proposes a comprehensive computational framework for evaluating the time-dependent reliability of underground water distribution pipeline networks subject to corrosion-induced deterioration. The framework integrates the stochastic variability and temporal evolution of corrosion rates, as well as the correlation of failure events across multiple paths. First, the reliability of individual pipelines is quantified using an analytical method that accounts for degradation over time. Then this is extended to the path level by evaluating connections from source to sink nodes. Finally, the network-level reliability is derived by incorporating interdependent path failures. The key contribution of this paper lies in addressing pipeline networks degraded by corrosion, where path interdependence significantly influences network-level reliability. The proposed path-based approach yields reliability estimates by explicitly modeling functional path interactions and joint failure behavior. Results presented in this paper indicate that increased path correlation enhances redundancy and reduces network failure probability. The developed framework offers a valuable tool for engineers and asset managers to support risk-informed decision-making and optimize maintenance strategies for corrosion-prone underground pipeline networks over their whole lifecycle.