Univariate SPC systems effectively only detect, or provide early warning of off-specification production, process disturbances and process malfunctions related to individual quality measurement sources. Consenquently, they provide litte information about the interactions between the variables which are so important in complex processes such as those now found in the process and manufacturing industries. These limitations can be addressed through the application of Multivariate Statistical Process Control (MSPC), The bases of MSPC are the projection techniques of Principal Components Analysis (PCA) and Projection to Latent Structures (PLS). Through the application of PCA or PLS, the process can be defined in terms of a much reduced set of latent variables (a linear combination of the original variables), which reflect the true dimensionality of the process This paper presents an overview of multivariate statistical process control for continuous and batch processes. The power of the methodology is demonstrated by application to two industrial processes.
Dong, D., and McAvoy, T.J.1994. 'Non-linear principal components analysis based on principal curve and neural networks', Proc ACC , Baltimore, USA, 1284-1288.
2.
Frank, P.M.1990. 'Fault diagnosis in dynatnic systems using analytical and knowledge-based redundancy — a survey and some new results', Automatica, 26, 459-474.
3.
Geladi, P., and Kowalski, B.R.1986. 'Partial least squares regression: A tutorial', Anal Chim Acta, 185. 1-17.
4.
Gertler, J.J.1988. Survey of model-based failure detection and isolation in complex plants, IEEE Control Systems Magazine, 8(6), 3 -11.
5.
Hastie, T.J., and Stuetzle, W.1989. 'Principal curves', J American Statistical Assoc, 84, 502-516.
6.
Hoskins, J.C., Kaliyur, K.M., and Himmelblau, D.M.1991. 'Fault diagnosis in complex chemical plants using artificial neural networks, CIChE J , 37, 137-141.
Hotelling, H.1947. 'Multivariate Quality Control'. In: Eisenhart, Hastay and Wallis (Eds) Techniques of Statistical Analysis, MacGraw Hill, New York, 111-184.
9.
Ignova, M., Glassey, J., Montague, G., Morris, A.J., and Kiparissides, C.1994. 'Neural networks and nonlinear SPC', 3rd IEEE Gonf on Control Applications, Glasgow, August, 1271-1276.
10.
Germarm, R.1984. 'Process fault detection based upon modelling and estimation methods — a survey', Automatica, 20, 387-404.
11.
Isermann, R.1995. 'Model based fault detection and diagnosis methods' , Proc American Control Conf, Seattle, USA.
12.
Joliffe, I.T.1986. Principal Component Analysis. Springer-Vedag , New York.
13.
Kaspar, M.H., and Ray, W.H.1992. 'Chemometrics methods for process monitoring and high performance controller design'. AIChE Journal, 38(10), 1593-1608.
14.
Kramer, M.A.1991. 'Non-linear principal component analysis using autoassociative neural networks', AIChE J, 37, 233-243.
15.
Kresta, J., MacGregor, J.F., and Marlin, T.E.1991. 'Multivariate statistical monitoring of process operating performance', Canadian J Chemical Engineering, 68(1),
16.
MacGregor, J.1994. 'Statistical process control of multivariate processes' , Preprints of the IFAC ADCHEM'94 Conf on Advanced Control of Chemical Processes, May, Kyoto, Japan.
17.
MacGregor, J.F., and Kourti, T.1995. 'Statistical process control of multivariate processes' , Control Eng Practice, 3(3), 403-414.
18.
MacGregor, J.F., Jaeckle, C., Kiparisides, C., and Koutoudi, M.1994a. 'Process monitoring and diagnosis by multiblock PLS methods ', Process Systems Engineering, 40(5), 826-838.
19.
MacGregor, J.F., Marlin, T.E., Kresta, J.V., and Skagerberg, B.1991. 'M ultivariate statistical methods in process analysis and control'. Proc of the Chemical Process Control (CPC-IV) Conf, South Padre Island, Texas, 18-22 February, AIChE, New York, Publication No P-67, 79-99.
20.
MacGregor, J.F., Nomikos, P., and Kourti, T.1994b. 'Multivariate statistical process control of batch processes using PCA and PLS', Preprints of the IFAC ADCHEM'94 Conf on Advanced Control of Chemical Processes, May, Kyoto , Japan.
21.
Martin, E.B., and Morris, A.J.1995. 'A new non-parametric multivariate confidence bound ', Internal Report, Department of Chemical and Process Engineering, University of Newcastle, UK.
22.
Martin, E.B., Morris, A.J., and Papazoglou, M.1995. 'Confidence bounds for multivariate process performance monitoring charts', IFAC Workshop on Fault Detection and Supervision in the Chemical Process Industries, Newcastle , UK, 33-42.
23.
Miller, P., Swanson, R.E., and Heckler, C.F.1993. 'Contribution plots: The missing link in multivariate quality control', 37th Annual Fall ConferenceASQC, Rochester, NY.
24.
Nomikos, P., and MacGregor, J.F.1994. 'Monitoring of batch processes using multiway principal components analysis', AIChE Journal,40(8), 1361-1375.
25.
Nomikos, P., and MacGregor, J.F.1995. 'Multivariate SPC charts for monitoring batch processes '. Technometrics, 37, 41-59.
26.
Papazoglou, M.C.1995. 'Multivariate statistical process control of batch processes using multiway principal components analysis: Overview and Application' , Internal Report 07/1995/1, University of Newcastle, Department of Chemical and Process Engineering, UK.
27.
Patton, R.J., Frank, P.M., and Clark, R.N.1989. Fault diagnosis in dynamic systems: theory and applications , Prentice Hall.
28.
Peel, C., Morris, A.J., and Kiparissides, C.1993. 'Process condition monitoring and neural network feature detection', Proc IFAC World Congress, Sydney, Australia, Vol IV, 65-68.
29.
Qin, S.Z., and McAvoy, T.J.1992. 'Nonlinear PLS modelling using neural networks' , Computers and Chemical Engineering, 16(4), 379-391.
30.
Saunders, A.C.1992. 'Process monitoring and nonlinear observers', PhD Thesis, Department of Chemical and Process Engineering, University of Newcastle, UK.
31.
Saunders, A.C.G., Morris, A.J., and Martin, E.B.1995. 'A contribution to nonlinear PLS in process modelling ', submitted to Computers and Chemical Engineerirg.
32.
Venkatasubramanian, V., and Chan, K.1989. 'A neural network methodology for process fault diagnosis ', AIChE J, 35(12), 1993-2002.
33.
Wetheral, W.B.1991 'Statistical process control for the process industries '. In: Frontiers in Statistical Quality Control, 4, Physica Verlag, Wurtzberg.
34.
Wise, B., and Gallacher, N.1995. 'Statistical process control of multivariate processes' . IFAC Workshop on Fault Detection and Supervision in the Chemical Process Industries, Newcastle. UK, ISBN 0-08-042607-7, pp. 1-20.
35.
Wise, B.M., and Ricker, N.L.1991. 'Recent advances in multivariate statistical process control: Improving robustness and sensitivity', IFAC Symposium, ADCHEM'91 Proceedings, Toulouse, France, 125-130, Pergamon Press.
36.
Wold, S.1978. 'Cross-validatory estimation of the number of components in factor and principal components models', Technometrics , 20(4), 397-405.
37.
Wold, S.1992. 'Non-linear partial least squares modelling. II. Spline inner function', Chemometrics and Intelligent Laboratory Systems , 14, 71-84.
38.
Wold, S., Kettaneh-Wold, N., and Skagerberg, B.1989. 'Nonlinear PLS Modelling', Chemometrics and Intelligent Laboratory Systems, Elsevier Science, Amsterdam, 53-65.
39.
Zhang, J., and Morris, A.J.1994a. 'Process fault diagnosis using fuzzy neural networks ', Proc American Control Conf '94, Baltimore, USA, June, 971-975.
40.
Zhang, J., and Morris, A.J.1994b. 'Fault diagnosis of a CSTR using fuzzy neural networks ', Proc IFAC Symp on Artificial Intelligence in Real Time Control-AIRTC'94, Valencia, Spain, October, 191-196.
41.
Zhang, J., and Morris, A.J.1995. 'Neural networks for non-linear systems modelling' , IEE Proc Control Theory Appl, 142(6), 551-561.
42.
Zhang, J., Martin, E.B., and Morris, A.J.1995. 'Nonlinear PCA and accumulated score plots', Internal Report, Department of Chemical and Process Engineering, University of Newcastle, UK.
