Statistical Control of Multivariant Processes through the Artificial Neural Network Multilayer Perceptron and the MEWMA Graphic Analysis

Abstract

Nowadays, the supervision of the quality in the production systems is a very important task. In order to reach this goal is necessary that quality of the products are supervised from a global approach where all the quality characteristics are taken into account. In addition, it is necessary to resort to multivariate statistical process control techniques to guarantee a correct control of the manufacturing system. A strong disadvantage of multivariate statistical control is the lack of a complementary analysis that interprets the presence of variation or signals out of control to locate the source that generates the instability in the manufacturing process. To give a solution to the lack of interpretation, this research proposes using an artificial multilayer perceptron neural network as an analysis mechanism for each out-of-control signal that is detected by the MEWMA multivariate control graphic, which is able to identify small changes in the quality of any process. The monitoring and control procedure was used in a process of bearings manufactured for the automotive industry, the results show that the mechanism detected and interpreted from small variations to significant changes, identifying the origin of the variation generated in the productive system with an accuracy of 99.86%.