A Study Of Exponentially Weighted Moving Average (EWMA) Methodologies

Loading...
Thumbnail Image
Date
1993-12
Authors
Lee, Toh Tong
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This thesis concerns the use of the EHHA chart, an alternative to the Shewhart and Cumulative Sum (CUSUM) charts, for detecting a change in a process. We have written this thesis with two objectives in mind: i) to review the control charting methodologies of the EWMA control chart, and ii) to suggest some additional enhancements that can further enhance the EWMA chart. In this thesis, we provide an overview of the EWMA charts for the monitoring of the process mean and the process variance. Enhancements to the EWMA given by various authors for b~th the process mean and the process variance are also discussed. While reviewing all the different enhancements proposed by various authors, we will compare each enhancement to the others which serve similar purposes and try to summarize the similarities and differences between these methods. The pros and cons of each of them will also be discussed. We also show how the EWMA chart can be constructed using the appropriate enhancements.Some general discussions on issues related to the EWMA chart are also given. The discussion on the advantages and weaknesses of the EWMA chart and a comparison of the EWMA chart with the CUSUM chart will enable us to decide whether or not to use the EWMA chart. We also include a brief discussion on the use of the EWMA for a nonstationary process, some possible directions for future research, and current developments and prospects for the EWMA chart. In addition to a review of some enhancements suggested by various authors, we have also given some suggestions of our own. For example, the plotting of residuals in conjunction with an EWMA chart to detect large shifts in the process mean or more effective any other abnormality, a method which is in detecting a shift in the process mean which causes a eduction in the dispersion, and a method for determining when changes in the process mean start occurring and the magnitudes of such changes.
Description
Keywords
Exponentially Weighted Moving Average (EWMA)
Citation