A joint monitoring scheme for automatically controlled processes

Abstract

It has been proposed to integrate statistical process control techniques with Automatic Process Control (APC) or engineering process control schemes to ensure continuous quality improvement. The data streams collected from an APC-controlled process have been shown to be critically dependent on the process settings. This paper proposes a univariate approach which combines the information provided by both the process output and input control actions based on a uniformly most powerful test. The proposed control chart can be customized to be sensitive to large or small shifts in a linearly-controlled process with an optimal efficiency in either transient or steady-state conditions. The presented numerical study is restricted to consider only APC-controlled stationary disturbances.

Acknowledgements

The author thanks two anonymous referees for their constructive comments which improved the content and presentation of this paper. This research has benefited from many valuable discussions with Professor Kwok-Leung Tsui and is partially supported by NSF-DMI grant 0200224.