APPLICATION OF MULTIVARIATE CONTROL CHARTS FOR MONITORING AN INDUSTRIAL PROCESS

Authors

  • Custodio da Cunha Alves
  • Elisa Henning
  • Robert Wayne Samohyl
  • Andrea Cristina Konrath
  • Olga Maria Carvalho Formigoni Walter

DOI:

https://doi.org/10.17058/tecnolog.v17i2.3636

Keywords:

Statistical Process Control, MCUSUM, Hotelling T2, MEWMA, Monitoring

Abstract

The effective simultaneous monitoring of the many quality characteristics of a production process often depends on statistical tools that are becoming more and more specific. The goal of this paper is to investigate, via an industrial application, if there are significant differences in sensitivity between the use of Multivariate Cumulative Sum (MCUSUM), Multivariate Exponentially Weighted Average (MEWMA) control charts and Hotelling T2 charts in the detection of small changes in the vector of means of the process. In doing this study, we used real data from a machining process. A MCUSUM control chart was applied to monitor the two quality characteristics of this process simultaneously. A MEWMA chart was also applied. The result was compared to the application of the Hotelling T2 chart, which showed that the MCUSUM and MEWMA control charts detected the change sooner. This study was fundamental in defining the best choice between the three charts for the multivariate statistical analysis of this industrial process.

Downloads

Download data is not yet available.

Published

2013-12-18

How to Cite

Alves, C. da C., Henning, E., Samohyl, R. W., Konrath, A. C., & Walter, O. M. C. F. (2013). APPLICATION OF MULTIVARIATE CONTROL CHARTS FOR MONITORING AN INDUSTRIAL PROCESS. Tecno-Lógica, 17(2), 101-107. https://doi.org/10.17058/tecnolog.v17i2.3636

Issue

Section

Sistemas e Processos Industriais