Vulnerability assessments on electric power infrastructure are critical stages are necessary to understand the effects of disruptions on system’s resilience. The infrastructures are modeled as complex networks (scale-free, power law, random), where there is uncertainty regarding whose networks components would be in an operational state. Exact calculation of performance metrics based on connectivity at the system level, considering component failures, face the burden of intractability, so appropriate approximations to exact metrics are greatly desirable. In this study, computational experiments where designed for generation of random networks to evaluate a selected number of approximated performance metrics for accuracy and efficiency. A Java code was developed for complex network generation and performance metrics calculation. The code allows different network topology options to choose from in order to evaluate and compare the approximated performance metrics.