Abstract
Using a normalizing transformation to monitor process dispersion has received a great deal of attention. The exponentially weighted moving average (EWMA) control chart based on the logarithmic transformation of sample variance for monitoring increases in process variability is one such dispersion chart. The traditional EWMA-type dispersion chart resets the EWMA statistic to zero whenever it falls below zero. This paper proposes a new EWMA dispersion chart (NEWMA) by truncating negative normalized observations to zero in the traditional EWMA statistic. The comparison result shows that the NEWMA chart outperforms the traditional EWMA chart for detecting dispersion changes, especially at small changes.
Additional information
Notes on contributors
Lianjie Shu
Dr. Shu is an Assistant Professor in the Faculty of Business Administration. His email address is [email protected].
Wei Jiang
Dr. Jiang is an Assistant Professor in Department of Systems Engineering & Engineering Management. His email address is [email protected].