A Note on Individual and Moving Range Control Charts
Raid Amin, Ronald A. Ethridge
Abstract
Raid Amin, Ronald A. Ethridge
Abstract
In some process control applications it may be necessary to limit the sample size to one measurement. A possible option would be to use an individual measurement chart for monitoring the process center, and using a moving range chart for monitoring process variability (X-MR). There has been an ongoing debate, and there are conflicting views in publications whether it is necessary to use a Shewhart moving range chart when there is already a Shewhart individual measurement chart being utilized. In this article it is shown that there is no disadvantage in using the X-MR procedure based on average run length values. However, there are other considerations that justify the use of the X-MR procedure over the X-chart alone. In this article we hope to offer some clarifications on this topic, and provide graphs that simplify the design of such charts. We limit discussions to the case where it is reasonable to assume normality and negligible autocorrelation in the data.
OpenAlex reports 33 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In some process control applications it may be necessary to limit the sample size to one measurement. A possible option would be to use an individual measurement chart for monitoring the process center, and using a moving range chart for monitoring process variability (X-MR). There has been an ongoing debate, and there are conflicting views in publications whether it is necessary to use a Shewhart moving range chart when there is already a Shewhart individual measurement chart being utilized. In this article it is shown that there is no disadvantage in using the X-MR procedure based on average run length values. However, there are other considerations that justify the use of the X-MR procedure over the X-chart alone. In this article we hope to offer some clarifications on this topic, and provide graphs that simplify the design of such charts. We limit discussions to the case where it is reasonable to assume normality and negligible autocorrelation in the data.
Key concepts: Control chart, Shewhart individuals control chart, \bar x and R chart, Control limits, X-bar chart, Chart, Range (aeronautics), Statistics