A critical look at some point-process models for repairable systems

The use of a modeldriven approach to the analysis of repairablesystems is considered and shown to be useful as a way of understandingthe characteristics of such a system. However, considerablestatistical problems arise from the use of a set of standardmodel-building elements. In particular, identification problemsarise in many of the models. The argument is illustrated byexamples from software reliability and mechanical reliability.The conclusion is that, in many cases, the exploratory data-analysisapproach is as effective as the use of more sophisticated models.