Phase II Monitoring of Logistic Regression Profile in Two - stage processes

In most of advanced processes, quality of a final product depends on the several quality characteristic in the previous stages. This is called a cascade property in multi-stage processes. On the other hand, sometimes the quality of a process or product is characterized by a relationship between a response variable and one or more explanatory variable(s).This relation is called profile. In some applications, the quality of a product is following binomial distribution. In these cases the logistic regression is used to model the relation between response variable and explanatory variables. In this paper, the monitoring logistic regression profile in a two stage process is investigated. The effect of cascade property on the performance of common control charts is studied. Finally two methods including T based control chart and Skinner’s residual based control chart are proposed for monitoring logistic regression profiles in a two stage processes in Phase II. The performance of the proposed methods is compared through simulation studies in terms of average run length criterion. The results show the suitable performance of tow proposed methods. Keywords-Two-stage process; Logistic regression profile; Skinner's residual