Construct the system of profile monitoring using concept of EWMA control chart  

Oleh:

Jen, Chih-Hung ; Fan, Shu-Kai S. ; Chen, Quan-Chi

Jenis: Article from Proceeding
Dalam koleksi: The 14th Asia Pacific Industrial Engineering and Management Systems Conference (APIEMS), 3-6 December 2013 Cebu, Philippines, page 1-8.
Topik: Statistical process control (SPC); nonlinear profile; change point; average run length (ARL)
Fulltext: 1291.pdf (330.95KB)

Isi artikel

In most statistical process control (SPC) applications, the process is typically assumed that the quality characteristic under study can be properly characterized by a single variable. However, in some particular circumstances, the quality-related response relies on the functional form between two or more variables. To deal with such a situation, a linear (or nonlinear) function is applied so that the relationship between the independent variables that can be appropriately characterized. This relationship is also called a “profile”. This paper proposed a new monitoring framework for the vertical density profiles (VDP) data, which used SIC and AIC criteria for selecting the change points. Using these change points, the mixture second-order models are used to piece-wisely approximate the VDP profile data. Next, the proposed EWMA4 control charts are applied to monitor the parameters in the mixture second-order models. The experimental results show that the proposed framework presented better performances via detecting outlying profiles and the out-of-control average run length (ARLOUT) test.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali