Development and Evaluation of Control Charts Using Exponentially Weighted Moving Averages

Abstract

Some studies have been done to construct quality control charts using the Exponentially Weighted Moving Average, EWMA. However, they have been done implicitly for subgroups of size n and individual data. Moreover, the emphasis of the control chart of the EWMA has been on the monitoring of only the process mean. Even though there are articles discussing the derivation of control charts of the EWMA to monitor process dispersion, their derivations are rather difficult and not easy to comprehend. In this paper, methodologies to construct control charts using the EWMA of a sample statistic for subgroups of size n and individual data are discussed explicitly and separately. A consistent and systematic approach for deriving control charts of the EWMA to monitor the process mean and dispersion is presented. The performance of these EWMA control charts is then compared to Shewhart X̄ and R charts using the average run length (ARL) as determined by computer simulation.

• Exponentially Weighted Moving Average
• Average Run Length

Additional information

Notes on contributors

Chern Hsoon Ng

Mr. Ng is a Ph.D. student in the School of Industrial Engineering and Management.

Kenneth E. Case

Dr. Case is Regents Professor of Industrial Engineering and Management. He is a Fellow of ASQC.