Designing accelerated life tests for generalised exponential distribution with log-linear model

摘要

This paper discusses the optimal accelerated life test designs for Generalised Exponential (GE) distribution with log-linear model under periodic inspection and Type I censoring. For shape parameter, design and high test stresses, the accelerated life test is optimised with respect to the low test stress and the proportion of test units allocated to the low test stress. The asymptotic variance of the maximum likelihood estimator of log mean life or qth quantile at the design stress is derived as an optimality criterion with equally spaced inspection times and the optimal allocation of units for two stress levels are determined. Results show that the asymptotic variance at the design stress is insensitive to the number of inspection times and to misspecifications of guessed failure probabilities at design and high test stresses. Procedures for planning an accelerated life test, including selection of sample size, have been discussed through an example.