Premium
Developing identification techniques with the integrated use of SPC/EPC and neural networks
Author(s) -
Shao Yuehjen E.,
Chiu ChihChou
Publication year - 1999
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/(sici)1099-1638(199907/08)15:4<287::aid-qre251>3.0.co;2-5
Subject(s) - identification (biology) , artificial neural network , computer science , data mining , artificial intelligence , biology , botany
Recently, a great deal of research has focused on integrating statistical process control (SPC) and engineering process control (EPC). Most of these studies have concluded that the integrated use of both SPC and EPC is superior in performance to the use of either alone. However, the majority of these studies have assumed that the assignable causes of a disturbance can be identified and removed as soon as the out‐of‐control signal is triggered by SPC. In practice, the identification of the assignable causes of a disturbance is not so straightforward. Using SPC and EPC control schemes, this study introduces a simple graphical aid technique to display the pattern of the underlying disturbance. Using EPC and a neural network (NN) scheme, this study focuses on the development of another technique to identify the nature of the assignable causes of the underlying disturbance. The effectiveness and the superiority of the proposed approaches are demonstrated through a series of simulations. Copyright © 1999 John Wiley & Sons, Ltd.