Design of accelerated degradation test with tightened critical values under random coefficient degradation rate model

This paper presents accelerated degradation test plans considering adoption of tightened critical values. Under a random coefficient degradation rate and log–linear acceleration models, the asymptotic variance of an estimator for a lifetime quantile at the use condition as the optimization criterion is derived where the degradation rate follows a lognormal and Reciprocal Weibull distribution, respectively and then the low stress level and proportions of units allocated to each stress level are determined. We also show that the developed test plans can be applied to the multiplicative model with measurement error.