Identifying change-point in polynomial profiles based on data-segmentation. Issue 4 (21st April 2017)

• Record Type: Journal Article
• Authors: Nie, Bin; Du, Mengying
• Is Part Of: Communications in statistics. Volume 46:Issue 4(2017)
• Page Start: 2513
• Page End: 2528
• Publication Date: 2017-04-21
• Subjects: Diagnostic -- Local change -- Nonlinear profile -- SPC; 00A06; Mathematical statistics -- Periodicals; Mathematical statistics -- Data processing -- Periodicals; Digital computer simulation -- Periodicals; 519.5
• DOI: 10.1080/03610918.2015.1053917
• Languages: English
• ISSNs: 0361-0918

Abstract

ABSTRACT: Profile monitoring is the use of control charts for cases in which the quality of a process or product can be characterized by a functional relationship between a response variable and one or more explanatory variables. Unlike the linear profile's simple structure, the nonlinear profile has relatively less attainment because of high complexity. Regression model is the initial method to analyze the phase I of nonlinear profiles, but it lacks sensitivity for local characteristic changes. This article presents a strategy comprising two major components: data-segmentation, to concisely detect the location of local change by overlaying grid points onto horizontal axis, and change-point detection via the maximum likelihood estimate. Simulated data set of a polynomial profile is used to illustrate the effectiveness of the proposed strategy, and is compared with Williams' T 2 multi-variable statistics.