Monitoring general linear profiles using simultaneous confidence sets schemes

doi:10.1016/j.cie.2013.11.014

Monitoring general linear profiles using simultaneous confidence sets schemes

2014-02-01

31

PRE

AI

DOI：10.1016/j.cie.2013.11.014

Original

Original request for help

Save

Abstract

En 中文

In this article we consider the quality of a process which can be characterized by a general linear profile. For monitoring the general linear profile, we mimic the charting scheme for the distribution of a univariate quality characteristic by using two individual charts for the mean and variance of the profile, respectively. For monitoring the mean of the profile, based on the concept of simultaneous confidence set we propose a novel exponentially weighted moving average (EWMA) chart, which takes the features of the entire profile into account. Then this chart is used together with an EWMA chart for the variance of the profile to monitor the whole profile. Simulation studies show the effectiveness and efficiency of the proposed monitoring scheme. Furthermore, a systematic diagnostic method in the literature is utilized to find the change point location and to identify the parameters of change in the process. Finally, we use an example from semiconductor manufacturing industry to demonstrate the implementation of the proposed monitoring scheme and diagnostic method. (C) 2013 Elsevier Ltd. All rights reserved.

Keywords:

Profile monitoring

Simultaneous confidence sets

EWMA

Average run length

Change point estimation

Phase II study

Monitoring general linear profiles using simultaneous confidence sets schemes

Abstract

Journal

Organization