Optimal design of double stresses accelerated life test with competing risks

摘要

In order to shorten test time and realize rapid evaluation of the reliability of high reliability and long lifetime competing risk products, based on the rules of optimum design plan of double stresses accelerate life test, this paper presents a method of simulation based on optimal designs for double stresses accelerated life test with competing risks (CRDS-ALT) and presents a procedure of the test with Monte-Carlo, the asymptotic variance estimation of 100pth percentile of the lifetime distribution of the product at use condition were considered as goal function. Through defining each test stress level and the test censoring numbers of corresponding stress as the design variables, we makes statistics of analysis with MLE theory, and establishes the optimal designed model of double stresses accelerated life test with competing risks based on simulation. Based on the interpolation fitting theory, the target function is presented, which reduces the emulation scale, improves the efficiency of the test, and provides a technical support for optimal design for AT in life prediction of electronic equipment.