Design of accelerated degradation testing with multiple stresses based on D optimality

Most products are affected by multiple stresses simultaneously, so it is necessary to study the optimal method for the accelerated degradation testing (ADT) with multiple stresses. This method is proposed for constant stress loading and step stress loading, respectively. Stochastic process is used to model the perfor-mance degradation of products. For the constant stress ADT (CSADT), uniform design and uniformly orthogonal design theory are used to determine the combined mode of diffident stresses, and for step stress ADT (SSADT), each stress is arranged in a step way. Under the constraint of the total experimental cost, the optimum problem is established with the objective that minimizing the variance of the estimation of model parameters based on D optimality, and optimal test variables including stress levels, sample size, testing time and the interval of performance inspection at each stress level are given. Finally, simulation examples are presented to illustrate the proposed method.