A Modified Monte Carlo Technique for Confidence Limits of System Reliability Using Pass-Fail Data

A Monte Carlo technique of Rice & Moore is modified to give a more accurate procedure for approximating confidence limits on system reliability using binomial component data. The modified lower confidence limit not only improves the Rice & Moore results but also compares well with most of the currently available methods for special series systems. Simulation results show that the modified method performs better in maintaining the nominal coverage probabilities than the original Rice & Moore procedure.