ABSTRACT
Parametric statistical methods for analyzing reliability data start with an assumption of the form of the life distribution. Once a life distribution is assumed, the statistical methods are used to obtain estimates for the parameters of the distributions–hence the terminology parametric methods. There are a large number of parametric reliability estimation methods. Only the principal methods are presented. The three principal methods that are discussed are: the graphical method; the method of moments; and the likelihood method. Most of the graphical methods are based on the general relationship between the cumulative hazard function and the reliability function. An important feature of the method of moments is that confidence intervals may be defined for most of the parameter estimates. The most widely used of the parametric techniques is the method of maximum likelihood. Under this very intuitive strategy, one selects as parameter estimates the values "most likely" to have produced the observed data.
