Number of Distinct Categories" or NDC Calculation in MSA Studies

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
With your "10 statistically significant buckets" I think you are maybe mixing up resolution and the notion of the increments versus tolerance against the way natural variation works...

Not really. What I am thinking of is not too different from the error in charting of "chunky data",.


Every observation you plot on a process behaviour chart has multiple sources of variation.

Yes, I am aware of the total variance equation.


In the context of the chart all we are interested in is can we pick up them shifts and patterns with whatever selection of western rules we choose.

Can you even see the shifts amid the gage error? And if so, can you see them clear enough to make decisions in a timely fashion?


Measurement error interacts with part variation in a process behaviour chart setting but we are only looking to pick up the patters and shifts so the notion of splitting out or dividing measurement error on its own isn't useful or required. In fact, if you do this for this purpose you will most likely condemn perfectly good measurement systems.

Actually, it is very necessary. Remember, every gage is bad, you have to have the evidence that it is good enough!

Of course, this doesn't mean that because we have an adequate ICC that these systems are adequate for sentencing good from bad parts, that's a different question.

A much easier one to discern. You only care about the specifications and the error about them.

ICC is not immune from inadequate estimates of part variation, no, its just a statistic and like all statistics it relies on its inputs. My significant point is that as a statistic the ICC is empirically proven to be an approach to portraying usefulness of the measurement process as opposed to the ndc.

I can see it may be more useful. Just need to wrestle its full usefulness. Dr. Wheeler generally has a good handle on his statistics. We have only disagreed on a few occasions...
 

Welshwizard

Involved In Discussions
I would be interested in your views on ICC when you have had time to read the papers and perhaps applied it to your applications and real data. If you have alternate papers and methods of discerning usefulness I would be interested to read these and I'm sure Dr Wheeler too, always open to new approaches and ideas.

Cheers
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
I am more in the gaging end of the business these days, so it is definitely an opportunity to to do some evaluations.
 

Welshwizard

Involved In Discussions
Perhaps take some of your previous data where you applied your 10 significant buckets approach and compare your conclusions with ICC and report back to the Cove.

Cheers
 

leftoverture

Involved In Discussions
I plan to explore the ICC method further when time allows. We are extremely busy at the moment so not sure what exactly I'll be able to get to it, but I have plenty of data to try it on!
 
Top Bottom