This thesis examines the existing theories and applications of Multivariate Statistical Process Control, outlines areas of difficulty and proposes a new technique of multivariate process control chart with input-output relationship for optimal process control. The process control techniques developed up to the present time focused on the fast detection of out-of-control signals, and achieved considerable success in that respect. The techniques reported on multivariate process control thus far include extensions of univariate process control charts to their multivariate counterparts, ranging from classical Shewharts charts to modern Cumulative Sum Process Control charts. Alternative approaches in this area include Principal Component Approach, Partial Regression approach, Baysian modelling and sequential tests on detection of change points. Although each method has its own limitation, these new developments have significantly contributed to the achievement of a constant high quality of products and services. The techniques of process control are yet incomplete. They require continuous attention, as production and service technologies are being continuously developed.In particular, the level of automation, re-engineering of production processes and ever demanding economic optimality of technology demand the re-engineering of statistical process control. The CFM chart developed in this thesis would open the door to this area of science and lays a critical foundation for future research / Doctor of Philosophy (PhD)
| Identifer | oai:union.ndltd.org:ADTP/235805 |
| Date | January 1998 |
| Creators | Pemajayantha, V., University of Western Sydney, Nepean, Faculty of Commerce, School of Quantitative Methods and Mathematical Sciences |
| Source Sets | Australiasian Digital Theses Program |
| Language | English |
| Detected Language | English |
| Source | THESIS_FCOM_QMS_Pemajayantha_V.xml |
Page generated in 0.009 seconds