In recent years, there has been considerable interest on the part of scientist and engineers in effectively designing accelerated degradation test, which is very useful to assess product's life information with long-life and high-reliability properties. An effective optimal ADT design method can provide a reasonable arrangement of accelerated stress level setting, sample allocation and test time. However, existing researches can hardly optimize all the multiple decision variables simultaneously in an efficient way. In the current study, an effective stochastic process based constant stress-accelerated degradation test (CSADT) model is constructed, and then an optimization design method is proposed based on the CSADT model, where the multiple decision variables involving stress level, sample allocation and test time arrangement for each stress level can be simultaneously optimized based on genetic algorithm (GA) and Monte Carlo simulation. Meanwhile, a comprehensive sensitivity analysis method incorporating orthogonal experimental design, normalization and range analysis is constructed to illustrate the robustness of the optimal design procedure. Finally, an optimal CSADT design for self-regulating heating cables is presented, and comparative results demonstrate the rationality and effectiveness of the proposed method.

近年来，科学家和工程师对有效设计加速退化试验产生了相当大的兴趣，这对于评估具有长寿命和高可靠性特性的产品的寿命信息非常有用。有效的优化 ADT 设计方法可以合理安排加速应力水平设置、样本分配和测试时间。然而，现有的研究很难以有效的方式同时优化所有的多个决策变量。在目前的研究中，构建了一个有效的基于随机过程的恒定应力加速退化试验（CSADT）模型，然后提出了一种基于CSADT模型的优化设计方法，其中涉及应力水平的多个决策变量，基于遗传算法（GA）和蒙特卡罗模拟，可以同时优化每个应力水平的样本分配和测试时间安排。同时，构建了结合正交实验设计、归一化和极差分析的综合敏感性分析方法，以说明优化设计程序的稳健性。最后，提出了一种用于自调节加热电缆的优化 CSADT 设计，对比结果证明了该方法的合理性和有效性。