A correlated Bernoulli state-space subset partition simulation method for efficient network connectivity reliability analysis

Ensuring the reliable connectivity of modern physical lifeline networks is essential for their safe and continuous operation. However, the complexity arising from numerous network components, diverse failure modes, and potential interdependencies among component failures poses significant challenges for accurate and efficient connectivity reliability assessment. To address these challenges, this paper proposes a method based on correlated Bernoulli variables (CBV) integrated with a state-space subset partition simulation (SSPS) method, referred to as CBV-SSPS. This methodology, grounded in the principles of subset simulation (SS), enables the efficient identification of critical network state subsets that govern connectivity loss while explicitly accounting for statistical dependencies among component failures. The accuracy and computational efficiency of the proposed method is demonstrated through two representative lifeline networks. Results show that the proposed CBV-SSPS method effectively captures the influence of correlated component failures on network connectivity reliability and achieves highly accurate estimates with substantially reduced computational cost, confirming its practicality and robustness for network connectivity reliability analysis.