CUSUM charts display the cumulative sums of the deviations of the sample values from the target. In other words, CUSUM is a sum of the differences between sample values and the target. Use CUSUM charts for in-control processes to detect small shifts away from the target. CUSUM charts are suitable for processes in which it takes some time to produce a single item.

Minitab generates two kinds of CUSUMs:

·    One-sided. The upper CUSUM detects upward shifts in process level and the lower CUSUM detects downward shifts. This chart uses upper and lower control limits to determine when the process is out of control.

·    Two-sided. This chart uses a V-mask, rather than control limits, to determine when an out-of-control situation has occurred.

The sample data consists of:

·    Twenty observations randomly sampled from a normal distribution (m =4, s =1)

·    Ten observations randomly sampled from a normal distribution (m =3, s = 1)

This represents a small shift in the mean - a reduction by one standard deviation.

Data: Sample.MTW (available in the Sample Data folder).