Research articles
ScienceAsia 37 (2011): 62-68 |doi:
10.2306/scienceasia1513-1874.2011.37.062
Extended Deming's model and data mining approach for diagnosis management
Supot Soommata,*, Sanguan Patamatamkulb, Thamrong Prempridia, Manop Sritulyachota, Pijarn Ineurec, Surapan Yimmand, Larry Kleine
ABSTRACT: We develop a data mining approach and an extended Deming's management model to save diagnosis time in the slider process of the hard disk drive industry. The data mining approach consists of five mining algorithms, namely, the K-Mean clustering, the Kruskal-Wallis test, the multivariate chart, the association rules, and the continuity-based measurement. They provide an automatic diagnosis on manufacturing data to determine the defective process stages, machines, materials, and methods. The extended Deming's model provides a close-loop management of diagnosis. This analysis framework helps engineers to identify defective factors rapidly in order to deliver diagnosis results within an hour. Additionally, all results of extended Deming's management loop can be recorded and converted to be useful wisdom for effective manufacturing management.
Download PDF
13 Downloads 5685 Views
| a |
Department of Engineering Management, Faculty of Engineering, Vongchavalitkul University, Nakhonratchasima 30000, Thailand |
| b |
Department of Civil Engineering, Faculty of Engineering, Naresuan University, Phitsanulok 65000, Thailand |
| c |
Department of Six-Sigma, Seagate Technology, Nakhonratchasima 30170, Thailand |
| d |
Department of Industrial Physics & Medical Instrument, Faculty of Applied Science, King Mongkut's University of Technology North Bangkok, Bangkok 10800, Thailand |
| e |
Department of Failure Analysis, Seagate Technology, Minnesota 55435, USA |
* Corresponding author, E-mail: ssoommat@yahoo.com
Received 30 Nov 2009, Accepted 1 Nov 2010
|
