一次性设备不能重复使用,不能通过试验观测到其真实的寿命,只能观测到寿命区间.为了在短时间内获得更多的寿命信息,对寿命服从指数分布的一次性设备进行了简单的步加试验.引进EM算法,解决了矩估计、极大似然估计等传统的统计方法估计参数比较困难的问题.
It is well-known that the disposable equipment cannot be reused and its true life span cannot be observed through experiments except for its lifetime section. In order to gain more lifetime information in a short time, a simple step-stress accelerated life-test (SSALT) is undertaken in the experiment towards disposable equipment of the lifetime exponential distribution. Whereby the EM-algorithm is introduced to estimate the parameters of the model. This method turns out to be better than the traditional statistical methods like moment estimation and maximum likelihood estimation to solve the above problem.