基于混合验前分布的复杂装备测试性评估
Evaluation of Complex Equipment Testability Based on Mixed Prior Distribution

合理有效地利用验前信息是小子样评估中的关键问题。针对复杂装备小子样测试性评估问题，考虑了不同验前信息的可信度和重要度，提出了一种混合Beta分布的复杂装备小子样测试性评估方法。算例表明，在小子样前提下，传统统计方法得到的测试性置信下限过于保守；现有的可信度度量方法大都直接基于数据层，没有考虑验前信息的来源。混合Beta分布的复杂装备小子样测试性评估方法不仅考虑了验前信息的可信度，还对验前信息的重要度进行了分析，评估结论更加合理，具有较好的工程应用价值。
It is of great importance to useany prior information effectively and reasonably in the evaluation of small samples. Therefore, a new testability evaluation method based on mixed Beta prior distribution is presented, while considering both the credibility and the importance of prior information as well as the testability evaluation of complex equipment in small samples. The results show that, according to classical methods using small binomial samples, the lowerconfidence limits of product testability are conservative. Most of the measurements for the credibility of the prior information are based on data. The evaluation results are aggressive due to the missing sources of prior information. Thus, the conclusion is reasonable, and this method is promising for engineering applications.