Optimal Design Based On Masked Data System Life Test

In the field of engineering,reliability test is a test method which plays an important role in the reliability evaluation of products.The optimal design problems are all based on competitive failure data.However,failure data of high reliability and high-performance products are no longer simply competitive failure data.Therefore,we present a new design scheme for this kind of life data--the optimal design scheme based on masked data.In this paper,the optimal design scheme for life test under constant stress is studied.Given some basic assumptions of life test under constant stress,the likelihood function of life data is obtained.Then the maximum likelihood estimation and Fisher information matrix are obtained,and according to the d-optimal criterion and v-optimal criterion,the optimal design scheme of the constant stress life test is discussed.Finally,the normal stress life test scheme of two-component and three-component cases is discussed.Secondly,the optimal design scheme of accelerated life test under multi-stress is studied.Given some basic assumptions of accelerated life test and accelerated equation,the likelihood function of life data is obtained.Then the maximum likelihood estimation and Fisher information matrix of the accelerated life test parameters are obtained,and the optimal design scheme of the multi-stress accelerated life test is discussed according to the d-optimal criterion and v-optimal criterion.Finally,the scheme of multistress accelerated life test under two-component and three-component conditions is discussed,and the numerical simulation is carried out to obtain the optimal solution and the optimal value.