Abstract
The conventional practice of employing the moving range chart (MR chart) as a companion chart to the control chart for individual observations (X chart) has been recently questioned by some authors. We evaluate the joint performance of the X chart and the MR chart for normally distributed observations when there is a shift in the process variance. We conclude, on the basis of the relevant probabilities, that the successive differences used on the MR chart do contain useful information on shifts in the process variance. A simple supplementary diagnostic test for shift in the process variance is suggested. This test does not require the plotting of a separate chart and avoids the disadvantages of the MR chart.
Additional information
Notes on contributors
S. R. Adke
Dr. Adke is a retired Professor of Statistics and a University Grants Commission Emeritus Fellow.
X. Hong
Mr. Hong is a graduate student.