Real-time performance reliability evaluation method of small-sample based on improved Bootstrap and Bayesian Bootstrap

The failure distribution from traditional reliability methods reflects the average characteristics of the same kind of product,which can't be used for analysis real-time reliability evaluation and for small sample is more helpless and Bayes method is still in dispute,so this paper introduced improved Bootstrap method and Bayesian Bootstrap method to real-time performance reliability statistical analysis of small sample.The two methods extended the amount of samples by numerical simulation.First,introduced the basic idea,improved the traditional Bootstrap method and gave an algorithm for calculating the real-time expectation and variance of performance parameter,then analyzed the applicability of Bootstrap method and Bayesian Bootstrap method.After presented that arithmetic of real-time performance reliability modeling.In the end,gave an example to show the validity and practicality of these two methods.