Abstract This paper studies the performance evaluation of a power transmission system in terms of its network topology, where edges represent transmission lines and nodes represent subsidiary stations. The power transmission network is modeled as a stochastic-flow network (SFN) due to the possibility of failure, partial failure, and maintenance of edges (transmission lines). Furthermore, correlation poses a particular concern for such an SFN because the simultaneous failure of multiple components can dangerously degrade performance. We develop a method to measure the impact of correlated failures on network reliability, which is defined as the probability of demand satisfaction. Experimental results show that correlation may produce a significantly negative impact on network reliability, especially when there is a high level of network demand. Thus, the proposed approach captures the influence of correlation on network reliability and offers a method to quantify the utility of reducing correlation.