Bayesian estimation of reliability indexes for cold standby system under general progressive Type II censored data

摘要

Purpose - This paper aims to deal with the estimation of the empirical Bayesian exact confidencelimits of reliability indexes of a cold standby series system with (n + k - 1) units under the generalprogressive Type II censoring scheme.Design/methodology/approach - Assuming that the lifetime of each unit in the system is identicaland independent random variable with exponential distribution, the exact confidence limits of thereliability indexes are derived by using an empirical Bayes approach when an exponential priordistribution of the failure rate parameter is considered. The accuracy of these confidence limits isexamined in terms of their coverage probabilities by means of Monte-Carlo simulations.Findings - The simulation results show that accuracy of exact confidence limits of reliability indexesof a cold standby series system is efficient. Therefore, this approach is good enough to use forreliability practitioners in order to improve the system reliability.Practical implications When items are costly, the general progressive Type II censoring scheme isused to reduce the total test time and the associated cost of an experiment. The proposed methodprovides the means to estimate the exact confidence limits of reliability indexes of the proposed coldstandby series system under this scheme.Originality/value The application of the proposed technique will help the reliabilityengineers/managers/system engineers in various industrial and other setups where a cold standbyseries system is widely used.