Joint Research Conference

June 24-26, 2014

Inference on Cpm for Autocorrelated Data Contaminated with Measurement Errors

Abstract:

In this paper, we discuss the properties of the estimator of the process capability index for autocorrelated data in presence of measurement errors. The basic reason for this choice of the    subject matter is that in many industrial applications (especially those from the process industry), process data are often autocorrelated; and even with the most advanced measuring instruments measurement errors cannot be entirely ruled out. Hence, both autocorrelation and gauge imprecision need to be taken into consideration for correct assessment of process capability. In this study, we consider the case of a first order stationary autoregressive process and compare the statistical properties of the estimator of the index in the error case with those of the estimator in the error-free case.