Modified residual control charts for processes with changing variances

Statistical process control has paid less attention to the possibility that the variance of the process may change over time such as economic or social processes. The generalized autoregressive conditional heteroscedastic(GARCH) model here is put forward to fit processes with changing variances and the modified standard and EWMA residual control charts are adopted for monitoring the residuals from the fitting model for the target process. The developed approach shows more validity and rationality. Lastly, an example is provided to illustrate the application and the effectiveness of the proposed control charts.