正态分布产品的Bayes可靠性抽样检验方案

Bayesian Reliability Sampling Inspection Plan for Normal Distribution

在正态分布产品的指标值为正,总体方差已知的情况下,利用最大熵先验分布,研究指标值均值的可靠性抽样检验方案.通过指标值均值的后验分布,依据生产方和使用方所能承担的最大风险的限制确定检验的最小样本量.算例说明其效果比经典统计方法确定的样本量更小,并且其运算方法更加便捷.

Under the premise that index oval mean of normal distribution is positive and its total variance of normal distribution is known,the Bayesian reliability sampling inspection plan is made on the basis of the maximum entropy prior distribution.According to posterior distribution of index oval mean,the minimum sampling size is obtained in the light of maximum risk to producers and users.The example shows that the effect is smaller than the classical statistical method,and the method is more convenient.