多源验前信息之下Bayes可靠性估计

本文考虑存在多源验前信息的情况，以二项分布为例，首先把各种验前信息化成不同的约束条件，并运用最大熵准则推导出各种验前信息所对应的验前分布，然后将这些分布综合成最终的验前分布，最后根据系统的寿命试验数据得出可靠性参数的验后分布并进行了Ｂａｙｅｓ推断，文中给出了仿真实例以说明方法的有效性。

Bayes Reliability Estimation Using Prior Information of Multiple Sources

We develop a Bayes estimator for the binomial survival problbility when there exist multiple sources of prior information.To begin with,we change prior information into functional constraints,and acquire various priors according to different information applying the Principle of Maximum Entropy.Then,a mixture of these beta priors is taken as the combined prior.After that,Bayes theorem is used to obtain the final mixed beta posterior distribution from which the desired estimates are obtained.At last ,an example illustrates the method.