Evaluation of the Productive Process by Means of Control Charts in the Presence of Volatility

The main purpose of this research work is to verify the stability of a productive process in the presence of the effects of autocorrelation and volatility so that these characteristics may be captured by a joint forecast model which produces residuals to be used in a control chart. Also, the effects of these factors will be analyzed to verify the impact in the performance of the control chart. The autoregressive integrated moving average (ARIMA) model was used along with the autoregressive conditional heteroskedasticity model (ARCH) to deal with the autoregression and volatility present in the data. The process stability was analyzed by means of control charts applied to the residuals coming from the joint model. The AR (1) - ARCH (1) model shows that the use of an appropriate forecasting model brings significant contributions to the control chart performance.