Abstract: Fault detection rate/ fault isolation rate (FDR/FIR) evaluation result with low confidence level is a very important problem in testability integrated evaluation, which is caused by the “small sample” field usage data, in order to settle the problem, the Bayes inference on dynamic population-based FDR/FIR integrated evaluation model and method are set up. Firstly, With the multivariate Dirichlet distribution as the prior distribution and the prior parameter of Dirichlet distribution are solved by using module test data and expert experience data. Then, the FDR/FIR integrated evaluation model and method based Bayes inference on dynamic population are considered and launched, in which the “small sample, dynamic population” growth test data in development stage and “small sample” field usage data can all be fused actively, and that the model robustness is investigated by simulation method. The study shows that the FDR/FIR integrated evaluation by means of such model can produce the evaluation conclusion with high confidence level in a short field usage period or under “small sample” condition. What has been investigated can offer an important theory base and method for testability integrated evaluation of equipments.

Key words: Dirichlet distribution, Inference theory on dynamic population, Integrated evaluation, Robustness, Testability