Robust Designs of Step-Stress Accelerated Life Testing Experiments for ReliabilityPrediction

In this article, we discuss the optimal and robust designs for accelerated lifetesting (ALT) when a step-stress plan is performed. It is assumed that the time to failureof a product has a Weibull distribution with a log-linear life-stress relationship. We adopta generalized Khamis-Higgins model for the effect of changing stress levels. Taking intoaccount that the assumed life-stress relationship is possibly misspecified, we have derivedthe optimal stress changing time of the simple step-stress plans in order to minimize theasymptotic mean squared error of the maximum likelihood estimator for the reliability ofa product at the normal use stress level and at a pre-specified time. The optimal 3-stepstressplans with minimum asymptotic squared bias are also discussed.