An alternative method for estimating mean life of power system equipment with limited end-of-life failure data

Traditional statistical reliability analysis relies on failure data for a population of devices. If a complete data set is available, statistical reliability analysis can provide predictions, such as mean-time-to-failure for a particular device, percentages of devices that will fail at a particular time or before a particular age, a statistical distribution of failure ages, and other statistical measures of device failures. However, a typical population includes devices that have not yet failed. In reliability analysis, such populations are often denoted as ldquoright censored populationsrdquo. In this paper, we propose a new method to evaluate mean life of power system equipment with limited end-of-life failure data. The method is based on the generalized exponential distribution. This method can be used as an alternative to methods based on normal and Weibull distribution models.