One approach for monitoring autocorrelated data consists in applying a control chart to the residuals of a time series model. However, due to the so called 'forecast recovery', the response to a mean shift in the observed process can appear attenuated in the residual series, in particular, after a short transient phase. To try to overcome this problem, we suggest a simple modification of the standard residual multivariate exponentially weighted moving average (MEWMA) control chart which reduces the 'forecast recovery' effect. Comparisons, based on two real industrial process models, show that the proposed modification can enhance the ability of the MEWMA control chart to detect both small and medium mean shifts.
An Enhanced Residual MEWMA Control Chart for Monitoring Autocorrelated Data
CAPIZZI, GIOVANNA;MASAROTTO, GUIDO
2009
Abstract
One approach for monitoring autocorrelated data consists in applying a control chart to the residuals of a time series model. However, due to the so called 'forecast recovery', the response to a mean shift in the observed process can appear attenuated in the residual series, in particular, after a short transient phase. To try to overcome this problem, we suggest a simple modification of the standard residual multivariate exponentially weighted moving average (MEWMA) control chart which reduces the 'forecast recovery' effect. Comparisons, based on two real industrial process models, show that the proposed modification can enhance the ability of the MEWMA control chart to detect both small and medium mean shifts.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.