ENBIS: European Network for Business and Industrial Statistics

Phase II Monitoring of Binary Response Profiles

23 September 2009, 10:45 – 11:05

Abstract

Submitted by

Rassoul Noorossana

Authors

R. Noorossana , M. Dorri, K. Paynabar, and A. Saghaei

Affiliation

Industrial Engineering Department , Islamic Azad University - South Tehran Branch, Tehran, Iran

Abstract

Abstract
In some industrial applications, quality of a product or process can be well characterized by a linear or nonlinear relationship between a response variable and one or more explanatory variables. Such relationships are referred to as profile. This paper discusses the case where quality of a product or process could be well modeled by a binary response profile. Two approaches based on likelihood ratio test and Pearson residuals are developed to monitor the stability of the parameters of the logistic regression model. The performances of the two proposed approaches are investigated numerically using average run length criterion. Results indicate superior performance of the control scheme which is based on Pearson residuals and two separate exponentially weighted moving average control charts.

Key Words and Phrases: Statistical process control, Profile monitoring, Logistic regression, Likelihood ratio test, Binary response, Phase II control, Pearson residuals

Return to programme