ABSTRACT
This chapter describes methods for estimating the capability of a process from variable data. Such data are usually measurements obtained on a collection of items. The data to be analyzed are assumed to be a random sample from a normal distribution. The problem of estimating the proportion of nonconforming items is discussed, based on both observations collected as individuals and observations belonging to subgroups. Estimation of various types of capability indices is described, including methods for obtaining confidence intervals. A distinction is made between indices calculated from a short-term estimate of the process standard deviation, which measure capability, and indices calculated from a long-term estimate, which measure performance.
