I have the answer for you on, “Truth be told I don’t know where the “m” comes from” … In the process capability index statistic Cpm, “m” is the total number of study subgroups which go in the calculation of the Cpm statistic.

The traditional variable control charts are to monitor the “centralizing” and “dispersing” tendency of specific quality variable in the process through sampling. When we sample basis rational subgroups, the statistic of each subgroup is the “mean” and “range” or the “mean” and “variance” of observations from a subgroup.

In traditional variable control charts occasionally, however, because of production problems, or inspection mistakes will cause a difference in the sample numbers within each sub-group. If such situation occurs, the control limits of a traditional control chart will vary with the sub-group size (i.e., as the subgroup size increases, the control limits become narrower; as the subgroup size decreases, the control limits become wider apart). Nevertheless, there will be no such problem faced if we use the ‘PCI’ based control chart, which is based on Cpm statistic.

Cheers,
Gagandeep

Great post summarising an often underused or at least not well understood capability metric. This is a nice link explaining the metric further, with some good examples: http://tinyurl.com/2j3ylu

Rob