A Novel Simulation-Based Adaptive MEWMA Approach for Monitoring Linear and Logistic Profiles

Abstract

As a common approach in the development of control charts in Statistical Process Control (SPC), an industrial process is monitored with one or more quality characteristics using their corresponding distributions. Note though, modelling the quality characteristics through a relation between some independent and dependent variables is an alternative approach which is designated as profiles monitoring. This study proposes the integration of the adaptive approach to the conventional Multivariate Exponentially Weighted Moving Average (MEWMA) control chart to improve its detection ability in phase II application. The run length characteristics of the adaptive MEWMA chart are measured with the use of Monte Carlo simulations by which better performance of the proposed method than numerous existing competitors including the conventional MEWMA chart is indicated in monitoring linear and logistic profiles. Finally, a real-life example from semiconductor manufacturing is provided to demonstrate the implementation and superiority of the proposed adaptive MEWMA chart over the conventional MEWMA chart.