New methods for phase II monitoring of multivariate simple linear profiles

Abstract

In some statistical process control applications, the quality of a process or product is characterized by a function that relates a response variable to one or more explanatory variables, referred to as a “profile”. In some cases, multivariate simple linear profiles are required for effective quality modeling. In these profiles, there is a set of correlated response variables, regressed on an explanatory variable. There have been only a limited number of studies on multivariate simple linear profiles monitoring. In this paper, three new methods are proposed based on multivariate homogeneously weighted moving average (MHWMA) control chart to improve monitoring of multivariate simple linear profiles. The performance of the proposed methods is evaluated by simulated ARL metric. A comprehensive comparison is also conducted between the performance of the proposed methods and the existing ones. The results indicated that the proposed methods functioned very well under the various conditions considered. In addition, the practical application of the proposed control charts is also demonstrated using two real case studies.

Keywords:

• Average run length
• Control chart
• Multivariate linear profile
• Profile monitoring
• Statistical process control