Title :
Using combination charts to cocoa powder production process control
Author :
Lee Siaw Li ; Djauhari, M.A.
Author_Institution :
Dept. of Math. Sci., UTM, Skudai, Malaysia
Abstract :
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 T2 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.
Keywords :
control charts; food processing industry; food products; process monitoring; quality control; statistical analysis; statistical process control; F statistic; Frobenius norm statistic; Johor Bahru; Malaysia food industry; T2 statistic; Wilks statistic; cocoa powder production; combination chart; control statistics; process control; production process monitoring; Covariance matrix; Image color analysis; Monitoring; Powders; Process control; Vectors; multivariate normal process; process level; process variability; sensitivity;
Conference_Titel :
Statistics in Science, Business, and Engineering (ICSSBE), 2012 International Conference on
Conference_Location :
Langkawi
Print_ISBN :
978-1-4673-1581-4
DOI :
10.1109/ICSSBE.2012.6396610