int . j . prod . res ., 2000, vol . 38, no . 3, 625±637 Statistical monitoring and diagnosis of automatic controlled processes using dynamic PCA F. TSUNG y As manufacturing quality has become a decisive factor in competing in a global market, statistical quality techniques such as statistical process control ( SPC ) are becoming very popular in industries. With advances in sensing and data capture technology, large volumes of data are being routinely collected in automatic controlled processes. There is a growing need for SPC monitoring and diagnosis in these environments, but an eective implementing scheme is still lacking. This research provides an integrated approach to simultaneously monitor and diag- nose an automatic controlled process by using dynamic principal component analysis ( DPCA) and minimax distance classi®er. Through a step-by-step imple- mentation procedure, the proposed scheme is expected to have an impact on many manufacturing industries with automatic process control ( APC) or engin- eering process control ( EPC) . 1. Introduction The importance of statistical quality techniques for process monitoring and diag- nosis is well recognized in industry. Currently most competitive manufacturing com- panies are implementing them to varying extents. However, these quality techniques cannot be implemented eectively in many automatic controlled processes. This paper develops an integrated scheme for eective monitoring and diagnosis of auto- matic controlled processes. Many manufacturing processes are equipped with automatic process control ( APC) or engineering process control ( EPC ) for short-term variation reduction. However, statistical process control ( SPC) techniques are still needed to detect the out-of-control conditions and to remove their root causes for long-term process improvement. Conventional SPC techniques are usually applied to the process out- puts after automatic control, and are often ineective as the information contained in the APC is ignored. More discussions on the ineectiveness of conventional SPC monitoring can be found in Messina et al. ( 1996 ), and Tsung et al. ( 1999 ) . Also, systematic root cause diagnosis is usually a missing link in SPC implemen- tation when SPC charts detect an out-of-control condition. Tucker et al. ( 1993 ) pointed out the critical need for eective means of diagnosing the root causes of signals provided by SPC. They suggested that well-planned and well-executed data gathering practices can yield an invaluable base of information useful for diagnostic analysis during the conduct of APC/SPC integration. However, little research has International Journal of Production Research ISSN 0020±7543 print/ISSN 1366±588X online # 2000 Taylor & Francis Ltd http://www.tandf.co.uk/journals/tf/00207543.html Revision received May 1999. { Department of Industrial Engineering and Engineering Management, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, PR China; e-mail: season@ust.hk