Estimation of component mean lifetimes of a masked system using unclassified system life data

We consider a masked system, which consists of several components. Due to cost and diagnostic constraints, the component mean lifetimes might not be exactly known. We consider a system where if one of components fails, the whole system fails and suppose that the lifetimes of components have exponential distributions. Since the exponential distribution has a large variance, it is hard to find the component mean lifetimes. We propose an estimation method, which will generally find the component mean lifetimes. The results of a Monte Carlo simulation study are presented to demonstrate the favorable estimation of the mean lifetimes of components using unclassified system life data. The estimation produced by our method is better than that produced by the K-means algorithm.