Improving the Performance of Exponentially Weighted Moving Average Control Charts

Authors

  • Mu'azu Ramat Abujiya,

    1. Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, Johor, Malaysia
    Search for more papers by this author
  • Muhammad Hisyam Lee,

    Corresponding author
    1. Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, Johor, Malaysia
    • Correspondence to: Muhammad Hisyam Lee, Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor, Malaysia.

      E-mail: mhl@utm.my

    Search for more papers by this author
  • Muhammad Riaz

    1. Department of Mathematics and Statistics, King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia
    2. Department of Statistics, Quaid-i-Azam University Islamabad, Pakistan
    Search for more papers by this author

Abstract

A control chart is a graphical tool used for monitoring a production process and quality improvement. One such charting procedure is the Shewhart-type control chart, which is sensitive mainly to the large shifts. For small shifts, the cumulative sum (CUSUM) control charts and exponentially weighted moving average (EWMA) control charts were proposed. To further enhance the ability of the EWMA control chart to quickly detect wide range process changes, we have developed an EWMA control chart using the median ranked set sampling (RSS), median double RSS and the double median RSS. The findings show that the proposed median-ranked sampling procedures substantially increase the sensitivities of EWMA control charts. The newly developed control charts dominate most of their existing counterparts, in terms of the run-length properties, the Average Extra Quadratic Loss and the Performance Comparison Index. These include the classical EWMA, fast initial response EWMA, double and triple EWMA, runs-rules EWMA, the max EWMA with mean-squared deviation, the mixed EWMA-CUSUM, the hybrid EWMA and the combined Shewhart–EWMA based on ranks. An application of the proposed schemes on real data sets is also given to illustrate the implementation and procedural details of the proposed methodology. Copyright © 2013 John Wiley & Sons, Ltd.

Ancillary