Optimal Design of-the Accelerated Degradation Experiment by Monte Carlo Method and Genetic Algorithm

There are extensive applications of accelerated degradation test in predicting the lifetime distribution of highly reliable products. The precision of the estimation can be improved by optimizing the experimental design of the accelerated degradation test. However, the complexity of the analytical method prevents the optimization algorithm from extensive application. In this work, a two-step method, based on Monte Carlo simulation and multi-objective genetic algorithm, is presented to optimize the accelerated degradation test, where the degradation rate follows a lognormal distribution. Then, a numerical example is provided to illustrate the method. The result of simulation and sensitivity analysis shows the optimized sample allocation ratio is closely related to the random measurement error.