Using combination charts to cocoa powder production process control

This paper deals with cocoa powder production process which is multivariate in nature. The process is monitored based on individual observations. In order to monitor such process, in general, there are three different statistics available in the literature namely T 2 statistic introduced by Hotellin...

Full description

Saved in:
Bibliographic Details
Published in2012 International Conference on Statistics in Science, Business and Engineering (ICSSBE) pp. 1 - 5
Main Authors Lee Siaw Li, Djauhari, Maman A.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2012
Subjects
Covariance matrix
Image color analysis
Monitoring
multivariate normal process
Powders
Process control
process level
process variability
sensitivity
Vectors
Online AccessGet full text

Cover

More Information
Summary:This paper deals with cocoa powder production process which is multivariate in nature. The process is monitored based on individual observations. In order to monitor such process, in general, there are three different statistics available in the literature namely T 2 statistic introduced by Hotelling in 1947, Wilks' statistic introduced by Mason, Chou and Young in 2009, and F (Frobenius norm) statistic recently introduced by Djauhari in 2010. Those statistics have their own properties and limitations. Therefore, the use of one statistic alone might not be sufficient to describe the actual process. To have a better understanding about the actual process, in this paper, a combination of those three separate control statistics will be used to monitor the production process of cocoa powder at a Malaysia food industry located in Johor Bahru. A promising and interesting result will be reported and discussed.
ISBN:1467315818
9781467315814
DOI:10.1109/ICSSBE.2012.6396610