A Dispersion Control Chart

Abstract

The study proposes a Shewhart-type control chart, namely Q chart, based on inter-quartile range, for monitoring changes (especially of moderate and large amounts which is major concern of Shewhart-type control charts) in process dispersion assuming normality of quality characteristic to be monitored. The design structure of Q chart is developed and its comparison is made with those of well-known R and S charts and R _Q chart proposed by Rocke (1992). The comparisons have revealed superiority of the proposed Q chart as compared to R and S charts in terms of discriminatory power and resistance against non normality and as compared to R _Q chart in terms of relaxing the restriction for 25th and 75th ordered positions in the data to be integer values and in terms of detecting special causes in the absence and presence of outliers in the data.

Keywords:

• Control charts
• Inter-quartile range
• Normality
• Non normality
• Outliers
• Power curves
• Q chart
• R chart
• R _Q chart
• S chart
• Simulations
• Special causes

Mathematics Subject Classification:

• 62P30
• 62F03

Acknowledgments

The author is grateful to the Associate Editor and referees for their constructive comments and suggestions which helped improve this article. Special thanks to Prof. Dr. Ronald J. M. M Does, Managing Director, Institute for Business and Industrial Statistics, University of Van Amsterdam, The Netherlands, for his invaluable suggestions and guidance for the revision and improvement of this manuscript.