小样本下维修性评估方法

Method of Maintainability Evaluation Based on Small Samples

小子样维修性评估在装备评估中频繁出现,需要建立一种有效的评估方法。针对小子样维修性评估,分析了自助法、随机加权法,并建立一种具有更好评估效果的改进自助法。首先以设定定参数产生一组小样本的随机数据,然后依据自助法、改进自助法和随机加权法的计算模型,分别对产生的随机数据进行分析计算,得出模型下随机数据的参数的估计值。将估计的参数与设定参数进行比较,模型较优则估计的参数误差较小。

Maintainability evaluation with small sample size frequently appears in equipment evaluation, so it is necessary to establish an effective evaluation method. In view of the situation, we analyzed the bootstrap method and random weighting method, and improved a bootstrap method that has better evaluation result. Firstly, we produced a set of random data through setting preset parameter; then analyzed and calculated the random data to obtain the estimation value of random number in the model on the basis of computational modeling of the bootstrap method, the improved bootstrap method and the random weighting method. By comparing the estimated parameters and preset parameters, we found the parameter error is smaller if the modeling is better.