An efficient method for network connectivity reliability computation considering correlation of components

Network connectivity reliability is imperative to lifeline systems, such as water distribution systems, and transportation networks. However, the computation of network connectivity reliability is highly complicated due to the large number of binary-stated network components, the diversity of post-hazard network topologies, and the correlation among network component failures. This paper presents an efficient method, integrated with Monte Carlo simulation (MCS), for computing network connectivity reliability specifically for binary-stated component networks, taking into account failure correlations. In particular, a concise transformation method for correlated discrete Bernoulli distribution is developed, enabling the capture of critical safe-failed states of network components as well as the conversion into independent standard normal space. The proposed method is then combined with an efficient shortest path search method to determine whether the network is connected or not, facilitating the computation of connectivity reliability for large-scale and dense networks. The results of examples show that the proposed method is more accurate and efficient than existing methods, and it can also obtain the influence range of component failure correlation on connectivity reliability. For disaster response and management of large, dense lifeline networks, the proposed method has significant advantages in providing fast and accurate assessments.