Process capability
If the process were stable, could it hold 32 µin?
Open bars sit past 32 µin. The curve is a normal shape on this sample. Cpk 0.57 is the arithmetic. It is not a capability claim.
How this sheet sits with the others
Mean Ra is 24.54 µin, the sample standard deviation is 4.37, and the upper spec is 32. A naive Cpk is 0.57. That number is arithmetic on an unstable series. The reject chart is not stable, and the day at 9 is past its limit. Capability sits in the middle of the argument. It does not replace the histogram, the gage, or the control chart.
- What it makes visible
- The Ra mean, the sample spread, and the spec, with the arithmetic Cpk kept in its place.
- When it is useful
- After the control chart has been read, and only then as arithmetic.
- Who uses it
- Priya Shah’s readings, set against Ruth Adeyemi’s limit of 32 µin.
- What it takes
- The Ra mean, the sample standard deviation, and the upper spec. Stability comes from the reject chart, not from this arithmetic.
- What it leaves
- A naive Cpk of 0.57 that is not a capability claim.
- The underlying record
- The Ra mean and spread against the 32 µin limit only mean something if the reject series at Devon Cho's cell is stable, which is the record the histogram, the gage, and the control chart already hold.
Where these facts go next
What this sheet still leaves out
It cannot see the gage, and it cannot see any day the chart is out of limits.