Moving range

Measures how variation changes over time when data are collected as individual measurements rather than subgroups. It equals the range of two or more consecutive observations.  When data are collected as individual observations, you cannot calculate the standard deviation for each subgroup. In such cases, the average moving range and median moving range across all subgroups are alternative ways to estimate process variation. You can create a control chart of moving ranges to track process variation when you have individual observations.

For example, a department store keeps track of the number of seconds it takes for operators to respond to customer calls. For six consecutive calls, here are the response times: 22, 35, 40, 20, 10, and 15. To calculate moving range of length 2, take the absolute value of the difference between consecutive data points.

You may want to use moving ranges of different lengths if the data are cyclical. For example, if you collect quarterly data, you might use a moving range of length 4 to ensure that one observation from each season is included in the calculation. To do this, subtract the minimum value from the maximum value of four consecutive observations. If you want to calculate a moving range of length 4 for the above example, the first moving range value is 40 - 20 = 20.