Intelligent Process Control System for Quality Improvement by Data Mining in the Process Industry
The large amount of bulky and noisy shop floor data is one of the characteristics of the process industry. These data should be effectively processed to extract working knowledge needed for the enhancement of productivity and the optimization of quality. The objective of the chapter is to present an intelligent process control system integrated with data mining architecture in order to improve quality. The proposed system is composed of three data mining modules performed in the shop floor in real time: preprocessing, modeling, and knowledge identification. To consider the relationship between multiple process variables and multiple quality variables, the Neural-Network/Partial Least Squares (NNPLS) modeling method is employed. For our case study, the proposed system is configured as three control applications: feedback control, feed-forward control, and in-process control, and then applied to the shadow mask manufacturing process. The experimental results show that the system identifies the main causes of quality faults and provides the optimized parameter adjustments.
KeywordsProcess Variable Quality Variable Control Application Shop Floor Knowledge Identification
Unable to display preview. Download preview PDF.
- Boulet, B., Chhabra, B., Harhalakis, G., Minis, L., and Proth, J. M., “Cell controllers: Analysis and comparison of three major projects,” Computers in Industry, 16 (3), 239–254, 1991.Google Scholar
- Mitra, A., Fundamentals of Quality Control and Improvement, Macmillan: New York, 1993.Google Scholar
- Montgomery, D. C., Design and Analysis of Experiments, 4th ed., Wiley: New York, 1997.Google Scholar
- Stockburger, D. W., Multivariate Statistics: Concepts, Models, and Applications, http://www.psychstat.smsu.edu/, 1997.
- Taguchi, G., Introduction to Quality Engineering, Asian Productivity Organization: Tokyo, 1986.Google Scholar