摘要：In this paper, we propose an optimization for the simple step stress accelerated life test for the Fr′echet distribution under type I censoring. The extreme value distribu- tion has recently become increasingly important in engineering statistics as a suitable model to represent phenomena with extreme observations. One probability distribu- tion, that is used to model the maximum extreme events, is the Fr′ echet (extreme value type II) distribution. A log-linear relationship between the Fr′ echet scale parameter and the stress are assumed. Furthermore, we model the e.ects of changing stress as a cumulative exposure function. The maximum likelihoo d estimators of the model parameters are derived. By minimizing the asymptotic variance of the desired life estimate and the reliability estimate, we obtain the optimal simple step stress accel- erated life test. Finally, the simulation results are discussed to illustrate the e.ect of the initial estimates on the optimal values.
关键词：Fr′ ; echet distribution; log-linear relationship; maximum likelihood estimator; optimal ; design; reliability; step stress accelerated life test; type I censored data
