Under the step-stress partially accelerated life test,based on the test data,the maximum likelihood estimation and Bayes-ian estimation of the distribution parameter,acceleration factor and reliability function are obtained by the classical method and Bayesian method,respectively.According to the asymptotic theory of maximum likelihood estimation,the approximate confidence intervals of the distribution parameter and the acceleration factor are constructed.The failure times of the censored units are predicted by using the observed data,including the best unbiased predictor and the conditional median predictor.The mean values and average relative errors of various estimators are calculated by Monte-Carlo method,and the influence of sample size on estimation accuracy is discussed.Finally,a numerical example is analyzed using the method presented in the paper.
关键词
指数分布/步进应力部分加速寿命试验/极大似然估计/贝叶斯估计
Key words
exponential distribution/step-stress partially accelerated life test/maximum likelihood estimation/Bayesian estimation
维普
万方数据