Optimum design of accelerated storage test plan under multiple stresses

In allusion to the complexity of environmental stresses electronic products encountered in the process of storage and the phenomenon that single stress is being used to conduct accelerated storage test currently, an optimum design method of temperature and humidity step-down-stress accelerated storage test based on the cumulative damage method is proposed. The minimum asymptotic variance of product's medium life estimated value at normal working stresses is taken as optimum target, each accelerated stress level as design variable. Meanwhile, combined with maximum likelihood estimation theory for parameter estimation, the mathematic model for multiple stresses accelerated storage test plan's optimal design is built. When genetic algorithm is employed to solve the optimum design result of accelerated storage test plan under multiple stresses of a meter, it is determined that the scheme is best when k equals 5. To improve the credibility, the Monte Carlo simulation is employed to compare the volatility of each scheme about minimum asymptotic variance and mean minimum asymptotic variance, which further proves that the scheme is optimal when k equals 5. The optimized test plan can not only solve optimization design for small sample accelerated storage test, but also ensure the accuracy of evaluation, reduce test times, shorten test time and save cost.