Process capability analysis on autoregressive process

Process capability analysis is conducted assuming that the process under study is in statistical control and independent observations are generated over time. However, in practice it is very common to come across process which, due to their inherent natures, generate autocorrelated observations. An approach that has proved useful in dealing with autocorrelated data is to directly model the correlative structure with an appropriate time series model, use that model to remove the autocorrelation from the data. The object of this paper was to generating the tables, which show the lower and upper 90%, 95% and 99% confidence limits for Cp and Cpk, for various level of confidence and autocorrelated. We also will provide tables for the engineers/practitioner to use for their in-plant applications.