Abstract:

In this paper, the optimum test plan and statistical inference of 3-step step-stress accelerated life tests (SSALT) under progressive Type-I censoring is studied. It is assumed that the lifetime of a test unit follow a Weibull distribution with mean lifetime of a unit is a log-quadratic function of stress level. The maximum likelihood and Bayesian method are used to obtain the point and interval estimates of the model parameters under progressive Type-I censoring. The Bayes estimates are obtained using Markov Chain Monte Carlo (MCMC) simulation based on Gibbs sampling. The optimum 3-step SSALT plan under progressive Type-I censoring is developed by minimizing asymptotic variance of the maximum likelihood estimators of log of mean life at the design stress. Finally, the numerical study is presented to illustrate the proposed study.