Loading…

Monitoring of Multivariate Processes Through the Regression Adjustment Procedure Based on Artificial Neural Networks

In current production systems, organizations must control the total of the variables that have a correlated influence on the final quality of the products. For this task, multivariate statistical control is used. One of the most used multivariate control methods is the so-called regression adjustmen...

Full description

Saved in:
Bibliographic Details
Published in:Revista IEEE América Latina 2019-06, Vol.17 (6), p.1020-1028
Main Authors: Ruelas, E., Vazquez, J., Cruz, J, Baeza, R., Sanchez, J., Jimenez, J.
Format: Article
Language:eng ; spa
Subjects:
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In current production systems, organizations must control the total of the variables that have a correlated influence on the final quality of the products. For this task, multivariate statistical control is used. One of the most used multivariate control methods is the so-called regression adjustment whose scheme proposes that to monitor a final quality characteristic in a product, a multiple regression should be applied on the previous variables that define it in the process and the resulting residue will be controlled from a univariate control chart to detect changes in the variations, however, the utility of the procedure is directly related to the efficiency of the method used to perform the adjustment task between the dependent and independent variables. The present work proposed to use an artificial neural network to carry out the adjustment operation, the results showed a remarkable superiority of the proposed scheme in comparison with the current scheme (Least Squares) and adjustment methods proposed in the literature (LAD and M). The proposed multivariate control approach allows the evolution of the regression adjustment method with the incorporation of artificial intelligence.
ISSN:1548-0992
1548-0992
DOI:10.1109/TLA.2019.8896825