The definition and estimation of process capability indices are usually based on the assumption that the production process under investigation is characterized by a normal distribution; however, the case of nonnormal processes occurs frequently in practice, as, for example, in the semiconductor industry. This paper addresses the problem of defining and computing reliable estimates for process capability indices (and particularly for C/sub pk/) for nonnormal processes; in particular, a curve-fitting approach to the estimation problem is taken and the problem of providing confidence intervals for the estimates of PCI's is considered. Finally, some application examples are also presented.