When applied in the right manner, process capability calculations are crucial for manufacturing.
I have worked with mass-produced parts. They often number in the hundreds of thousands each year. Process capability analysis has been a key part of my engineering tasks in various roles.
One common analysis is Cpk, defined as:
Cpk = min (USL – μ / 3σ), (μ – LSL / 3σ)
Where:
-
USL is the upper specification limit
-
LSL is the lower specification limit
-
σ is the standard deviation
-
μ is the average
Cpk shows how well a process meets specifications. This is true even if the process mean isn’t centered between the limits.
The minimum Cpk value is 1.00. This means the process fits within tolerances, which allows for no room for drift. For existing processes, a typical value is 1.33.
Years ago, in a meeting, someone unfamiliar with Cpk calculations said that a production run of about 40 assemblies had a Cpk below 1.00. This implies the process is unsound.
The team tested all assemblies, and they always measured within tolerance.
The production sampling rate was 100%, yet calculations suggested it should not be. I explained that a low sampling rate, like 30 parts out of 100,000, can still be useful for Cpk.
With a 100% sampling rate, Cpk calculations lose their meaning. They might reject qualified parts or lead to unnecessary changes in a good process.
In this case, the engineers prevailed, and the process remained unchanged.
Norman T. Neher, P.E.
Analytical Engineering Services, Inc.
Elko New Market, MN
www.aesmn.org