基于Bayesian Bootstrap小样本产品性能可靠性评估  

Performance reliability assessment based on Bayesian Bootstrap method

作　　者：万让鑫[1] 吴西良[1] 

机构地区：[1]中国人民解放军73903部队

出　　处：《信息技术》2012年第5期174-176,180,共4页Information Technology

摘　　要：性能可靠性已经成为现代电子产品重要的质量指标。对于现代电子产品的性能可靠性评估,传统经典统计方法面临着数据不足的问题,而Bayes方法容易引起争议,为此将BayesianBootstrap方法引入到产品性能可靠性数据的统计分析中,提出了基于Bayesian Bootstrap仿真的性能可靠性评估方法。该方法又称随机加权法,是一种通过数字仿真技术扩大样本量的方法,它完全依赖于样品本身的数据,不需要任何主观假设。首先介绍了Bayesian Bootstrap方法的基本思想,给出了利用该方法计算性能可靠性参数的期望与方差的算法,并分析了其适用范围,最后通过具体实例验证了该方法的有效性和实用性。Performance reliability is an important quality index for much electronic equipment. For performance reliability assessment of modem electronic product, classic method always lacks test data and Bayes method is still in dispute, so this paper introduces Bayesian Bootstrap method to performance reliability statistical analysis. Bayesian Bootstrap method extends the amount of samples by numerical simulation. First, it introduces the basic idea and gives out an algorithm for calculating the expectation and variance of performance parameter, then analyses the applicability of Bayesian Bootstrap method. In the end, an example is presented to show the validity and practicality of this method.

关 键 词：性能可靠性 随机加权法 性能试验 小样本 

分 类 号：TB114.3[理学—概率论与数理统计]