成败型系统变总体下的Bayes可靠性试验鉴定方法

提出了一种成败型系统变总体下的Bayes可靠性试验鉴定方法.该方法依据产品研制过程中不同阶段的试验数据,建立了产品可靠性的动态参数模型,以此描述产品可靠性的变化趋势,并对正样阶段的产品可靠性进行预测,利用最大熵法给出了产品可靠性的验前分布,最后结合产品少量的现场试验数据,对正样阶段的产品可靠性做出了Bayes统计决策.通过算例对比分析,该方法很好地解决了小子样产品的可靠性试验鉴定问题.

Bayesian reliability test evaluation for binomial system In varying population case

A Bayesian reliability evaluation method for binomial system in varying population case is proposed. The proposed method builds a dynamic parameter model according to the multiple-stage sample from the product development, which is capable of describing the changing trend of product reliability and predicting the reliability at product-stage. Then, the prior distribution of reliability at product-stage is obtained by using the Maximum Entropy method, and the Bayesian statistical decision for reliability at product-stage is made by combining the prior distribution with small sample size field data. Finally, a numerical example is given to illustrate the efficiency of our method in solving reliability evaluation under the condition of small sample size.