ABSTRACT
In most of the modeling and analysis of warranty costs and related quantities in the previous chapters, stochastic elements were handled by means of probability distributions stated generically (e.g., “F(⋅)”) in the general expressions given, and explicitly, by means of a formula for a specific distribution (e.g., the exponential), in the examples. The resulting expressions invariably involved certain parameters or other characteristics of the distribution (e.g., the failure rate λ or the mean time to failure, µ = 1/λ, in the exponential). In application of the numerous formulas, it was assumed that the numerical values of these quantities were known. In practice, this is rarely the case, and it is usually necessary to estimate the values of these characteristics from sample data of various types. The purpose of this chapter is to provide a brief survey of some of the important statistical techniques that may be used to analyze data for the purposes of estimating warranty costs.
