多源验前信息下的Bayes信息融合方法

本文以指数分布寿命型电子产品为例,讨论了多源验前信息情况下如何对产品失效率进行融合估计的问题。首先,介绍了寿命试验数据的信息融合方法,并以此为例,说明了取伽玛分布为指数分布寿命型产品失效率的先验分布的合理性。然后,在产品具有多源验前信息源的情况下,给出了一种方便实用的Bayes信息融合方法,充分融合各验前信息,得到产品失效率的验前分布及后验分布。最后给出了数值实例。

Bayes Information Fusion in Multi-sources of Prior Information

In this paper a method for pooling failure rate data obtained from different sources is discussed, and for the case of the exponential life distribution product. First we review the combination of data from life tests, the consideration of this show that a Gamma distribution for the prior of failure rate can be assumed. Then give a conveniency and applied Bayes-information fusion method when prior information comes from different sources, and find the prior distribution and posterior distribution of the failure rate. An example is presented in the end.