屏蔽数据在新型截尾样本下系统的可靠性分析

当系统含有屏蔽数据时,在具有随机移走逐步增加型截尾模型下,讨论了部件寿命服从双参数指数分布的串联系统可靠性估计问题。设随机移走系统数服从二项分布,利用极大似然方法,Bayes理论及方法,推导出双参数指数部件参数、系统可靠性函数、失效率函数及移走概率的极大似然估计和Bayes估计。并利用Monte Carlo方法对两种估计结果进行了比较,表明Bayes估计较极大似然估计效果更优。

Reliability Analysis of System Based on New Type Censored Tests Using Masked Data

Reliability estimations of the two-parameters exponential components in series system based on progressively type-Ⅱ censored model with random removals are discussed when using masked data.Considering that the number of systems removed at each failure time follows a binomial distribution,the paper deduces the MLE and Bayes estimations of unknown parameters,probability of removals,reliability and hazard function of the system.Also,Comparison of the two estimation method is given from numerical simulation applying the Monte Carlo method,which indicates that Bayes estimate is better than MLE.At last,it discusses the effect of different probability of masking and removals to accuracy of estimation.