Effect of measurement error on joint monitoring of process mean and coefficient of variation

Abstract

A control chart is used to monitor longitudinal measurements of a quality characteristic and detects undesirable causes of variability. In practice, measurement error may exist and deteriorate the detection capability of a control chart. This paper focuses on the effect of measurement errors on the performance of a joint monitoring scheme for mean and coefficient of variation (CV), which is a useful measure to monitor variability when the process mean fluctuates over time and process standard deviation is a linear function of process mean. Our results from the simulations and real data application demonstrated that measurement error affects the performance of the joint monitoring control chart and multiple measurements are recommended to avoid the adverse effects of measurement error.

Keywords:

• Control chart
• coefficient of variation
• measurement error
• log-normal transformation
• EWMA
• average run length

MATHEMATICAL SUBJECT CLASSIFICATION:

• Applications of statistics in engineering and industry
• control charts (62P30)

Acknowledgments

The authors are grateful to the reviewers for their suggestions which helped in improving substantially, the earlier version of this article.