在对数正态分布场合下,针对定时截尾步进应力加速寿命试验和步降应力加速寿命试验,根据准则1对总的产品失效数的比较和准则2对产品条件平均寿命的比较,分别对两种试验方式的效率进行比较分析。而针对定数截尾步进应力加速寿命试验和步降应力加速寿命试验,根据准则3对总试验时间的比较和准则4对样本均值的比较,分别对两种试验方式的效率进行比较分析。并以简单步进应力加速寿命试验和步降应力加速寿命试验为例,通过大量的Monte-Carlo模拟,在四个不同的准则下分别比较两种试验方式的效率,可以得到步降应力加速寿命试验有利于提高试验效率,即步降应力加速寿命试验优于步进应力加速寿命试验。因此,认为步降应力加速寿命试验很值得推广。
Based on the lognormal distribution, the efficiencies for the type-Ⅰ censored step-up-stress and step-down-stress accelerated life tests are compared and analyzed according to the first criterion by comparing the total numbers of failure products and based on the second criterion by comparing the conditional expectations of life. Also for the type-Ⅱ censored step-up-stress and step-down-stress accelerated life tests, the efficiencies for these two tests are compared and analyzed according to the third criterion by comparing the total test time and based on the fourth criterion by comparing the sample means. Afterwards the efficiencies for the two kinds of tests under these four different criterions are compared by using a large amount of Monte-Carlo simulations in one example of simple step-up-stress and step-down-stress accelerated life tests. As a result of this example, we see that the step-down-stress accelerated life test is available to improve the test efficiency and it is superior to the step-up-stress accelerated life test. Therefore, it is worthwhile to popularize the step-down-stress accelerated life test.