Optimum Constant-Stress Accelerated Life Tests with Competing Causes of Failure under Exponential Distribution

This paper persents optimum plans of constant-stress accelerated life tests with comoeting causes of failure, wihich life disrribution of each failure cause is assumed to be exponential. We obtain the optimum test designs to minimize the sum of asymptotic variance the Maximum Likelihood Estimation （MLE） of the log mean life in each cause at a design stress. A series of interesting results are shown in this paper, which are similar to or idential with those results shown in the optimum plans of step-stress accelerated life tests.