This paper considers the optimal design of accelerated life tests (ALT) under Type II progressive censoring with random removals, where the number of units removed at each failure time follows a binomial distribution. The asymptotic variance of the estimated qth quantile is derived. Monte Carlo simulations are conducted to find the optimal stress levels and the proportion of allocation for different combinations of parameters, which minimize the asymptotic variance of the estimated qth quantile. The results are presented. Finally, some suggestions are given to experimenters for selecting an appropriate plan in designing an ALT under the proposed censoring scheme. |