Process Capability sixpack on a Profile

Lets look a few scenarios, and I will give you my . This is not based on any research or statistics, but simply by thinking the problem through rationally.

• Scenario 1, Maximum Flatness: Zero is a natural boundary and there is a maximum allowable flatness specification. Presumably, there is some incentive to drive for smaller and smaller levels of flatness, thus moving farther and farther away from the USL. In this situation, you will conceivably utilize the entire tolerance spectrum. This would tend to justify not entering a lower specification because the Cpk would penalize you as you approach zero.
• Scenario 2, Maximum activation force: In this scenario, you are not going to strive for zero activation force. If you strove for zero activation force on a conventional spring-loaded design, there would no force left to reopen the switch. Therefore, you will also “hover” a safe distance below the maximum force and not decrease over time. Again, you would not want to enter zero as a lower spec limit because you may not want to center the activation force.

 

Logic before math…who knew?

 

So my a two cents above would apply to Flatness and Perpendicularity. For many detailed discussions on Position, I recommend a search at our sister forum Elsmar Cove. This is one such discussion thread.

Regarding Profile, this is another animal entirely. With flatness and perpendicularity, you are taking a worst case value for each part you measure, so it's fairly easy to perform a capability analysis. But with Profile, you have a multitude of individual measurements for each part (theoretically, an infinite number). If the profile is an irregular surface as opposed to a diameter, you would have to perform an infinite number of capability studies for each portion of the surface. Even for a uniform surface such as a diameter, how would you quantify Profile as a single number? You could conceivably reduce it to a maximum and a minimum value and perform a capability analysis for each, but this would misrepresent the within part variation. The only way I can see you doing a Profile capability would be to select a specific location on the profile, collect the data and perform the capability study. However, you basically ignore the rest of the profile. And, you would enter both an upper and lower specification even if the callout is unilateral.

 

Are you calculating capability indices because of a Customer requirement, internal requirement or just becuase you think it’s a good thing to do?

One of the best ways to start thinking about your results is to plot the data on a run chart or multi-variable. Sometimes a histogram is sufficient.
For a profile I would plot the various ‘profiles’ on a ‘scatter diagram’ where the x axis is the location and the Y axis is the profile value. Each part would be a 2 dimensional line showing the profile. I would then add a spec line for each location…takes some thought but can be done in EXCEL. This way you can actually ‘see the profile…