Analysis of non-repairable cold-standby systems in Bayes theory

作者:Jia, Xiang*; Guo, Bo
来源:Journal of Statistical Computation and Simulation, 2016, 86(11): 2089-2112.
DOI:10.1080/00949655.2015.1101464

摘要

In this paper, we seek to analyse the reliability of k-out-of-n cold-standby system with components having Weibull time-to-failure distribution in view of Bayes theory. At first, we review the existing methods exhaustively and find that all these methods have not considered Bayes theory. Then we modify the simplest method and propose new methods based on Monte Carlo simulation. Next, we combine all the information to derive the posterior distribution of Weibull parameters. A robust and universal sample-based method is proposed according to the Monte Carlo Markov Chain method to draw the sample of parameters to obtain the Bayes estimate of reliability. The drawn samples are proved to be rather satisfactory. Conducting a simulation study to compare all the methods in terms of accuracy and computational time, we have presented some useful recommendations from the simulation results. These conclusions would provide insight on the application for k-out-of-n cold-standby system.