Performance Improvement Study of CV-EWMA Control Chart to Detect Small Shifts of CV

Abstract:

In order to control a process that has short production cycle and where the product type and specifications change often with conventional shewhart control charts such as and control charts, a new control chart must be applied every time the parameters change . As this is a very inefficient method in terms of the cost and time, CV control chart using coefficient of variation statistics was developed. As CV control chart reflects only the current sample data on control chart, it can be useful when there is a significant change in process. However, it does not respond sensitively to a process that has subtle change or requires a high control level. CV-EWMA control chart was researched to monitor small shifts in CV. This study proposes a way to improve accuracy and precision of population parameter estimation of conventional CV-EWMA control chart and applied it to a control chart before analyzing its performance. As a result, the accuracy and precision of conventional CV-EWMA control chart has been improved and it was verified that the proposed control chart is a proper control chart to control small shifts of CV.

• Periodical: Advanced Materials Research (Volume 1051)
• Pages: 1016-1022
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.1051.1016
• Online since: October 2014
• Authors: Young Soo Park*, Eui Pyo Hong, Kyoung Yong Park, Woong Hee Shon
• Keywords: Coefficient of Variation (CV), Control Chart, CV-Exponentially Weighted Moving Average (CV-EWMA), Exponentially Weighted Moving Average (EWMA)

References

[1] C. Kang, M. Lee, Y. Seong, D. Hawkins, A Control Chart for the Coefficient of Variation, Journal of Quality Technology. 39 (2007), 151-158.

DOI: 10.1080/00224065.2007.11917682

[2] D.C. Montgomery, Introduction to Statistical Quality Control, 7th Ed., John Wiley & Sons, New York, (2013).

[3] E. Hong, C. Kang, J. Baek, H. Kang, Development of CV Control Chart Using EWMA Technique, Journal of the Society of Korea Industrial and System Engineering. 31 (2008), 114-120.

[4] E. Hong, H. Kang, C. Kang, DEWMA Control Chart for the Coefficient of Variation, Advanced Materials Research. 201-203 (2011), 1682-1688.

DOI: 10.4028/www.scientific.net/amr.201-203.1682

[5] J.E. Connett, W.W. Lee, Estimation of the Coefficient of Variation from Laboratory Analysis of Split Specimens for Quality Control in Clinical Trials, Controlled Clinical trials. 11 (1990), 24-36.

DOI: 10.1016/0197-2456(90)90029-2

[6] L.S. Nelson, Some Notes on Variation, Journal of Quality Technology. 31 (1999), 459-462.

[7] S.W. Roberts, Control Chart Tests Based on Geometric Moving Average, Technometrics. 1 (1959), 239-250.

DOI: 10.1080/00401706.1959.10489860

[8] W. Reh, B. Scheffler, Significance Tests and Confidence Intervals for Coefficient of Variation, Computational Statistics and Data Analysis. 22 (1996), 449-452.

DOI: 10.1016/0167-9473(96)83707-8

[9] Y. Bao, Finite-Sample Moments of the Coefficient of Variation, Econometric Theory. 25 (2009), 291-297.

DOI: 10.1017/s0266466608090555