Application of neural‐based algorithms as statistical tools for quality control of manufacturing processes

• Published in: Mathematical and Statistical Methods in Food Science and Technology pp. 431 - 448
• Main Authors: Pacella, Massimo, Semeraro, Quirico
• Format: Book Chapter
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 03.03.2014
• Subjects: artificial intelligence techniques; food science; machine learning theory; network's architecture; neural networks; neural‐based algorithms; pattern recognition tasks; quality control; statistical process control implementation
• ISBN: 1118433688; 9781118433683
• DOI: 10.1002/9781118434635.ch22

This chapter targets issues on the use of neural networks (NNs) for quality control of manufacturing processes, concerning the way of operation of each network model, the network's architecture and the results provided. It gives a brief overview of machine learning theory. This is followed by a review of the literature on the general topic of NNs for the automation of statistical process control (SPC) implementation. Artificial intelligence techniques are promising tools for the automation of SPC implementation. The chapter also discusses applications of NNs for pattern recognition and for detection of mean and/or variance shifts in process. Some studies reported the use of NNs for mean process shifts in the presence of autocorrelation in the data. This problem has more influence on the foods, chemicals, papers and woods industries. It can be concluded that multilayer perceptron (MLP) algorithms are the most widely used NNs for the determination of mean and variance shifts in process.