Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistics

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

I have a SQE that is referanceing the AIAG manual that shows capability of 1.67 is required. It does not show if it is a Cpk or Ppk. He is holding me to a Ppk of 1.67 which is very tight.

Can anyone shed some light on which would apply?

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

Bradson said:

I have a SQE that is referencing the AIAG manual that shows capability of 1.67 is required. It does not show if it is a Cpk or Ppk. He is holding me to a Ppk of 1.67 which is very tight.

Can anyone shed some light on which would apply?

Click to expand...

Hello! Welcome to The Cove Forums!

What AIAG manual is your SQE referring to? PPAP? SPC?
Is this for an initial PPAP submission?

If this is what your customer wants and you contractually agreed to it, then you have to meet that requirement.

You may also want to do a search here on Cpk and Ppk.

Good luck!

Stijloor.

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

there used to be a requirement for 1.67 if your process wasn't in statistical control? It's been so long since I was in automotive I can't remember the specifics. maybe this will spark someone's memory...

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

Stijloor

Thanx for the input. He is taking his comments from the AIAG PPAP manual. I have looked at it closer and am going to discuss it with him further. The way I see that it reads in Note 1 on page 9 "Meeting the initial process study capability acceptance criteria is one of a number of customer requirements that lead to an approved PPAP.

Since they refer to capability acceptance rather than performence I intend to use this to support my argument that the manual refers to Cpk

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

A PPAP is normally provided for new or changed product.

If the product is new, you only have short term data available, so can only provide Cpk results.

If the product is changed, you should have long term data available for the unchanged characteristics, but may only have short term data available for changed characteristics. Unchanged characteristics would have Ppk available and changed characteristics would only have Cpk available.

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

OK I'm making a classic mistake and trying to remember the requirement. but I'll proceed anyway.

I think the requirement is that when you dont' have long term process data (multiple subgroups) the requirement is to calculate the 'short term' capability by using data from only one setup or raw material batch and calculating the standard deviation for that single large subgroup. So the standard deviation is Stotal instead of Swithin . Stotal is the calculation for Ppk and Swithin is the calculation for Cpk. So in this case, you are using the Ppk formula to calculate the short term (Cpk) value! Of course if you have multiple setups, raw material lots and different operators then you can use the Cpk formula to calculate the short term capability.

Still don't remember the 1.67 requirement tho. For some reason I'm thinking they are two separate issues...it woudl be helpful if someone could quote chapter and verse from the manual!

I also seem to remember that both issues have been discussed in this forum before in some detail, but my memory of the automotive world was erased long ago!

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

Nice

By

RENGARAJ..M

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

AJLenarz said:

I agree by reading the above posts that there appears to be some level of confusion. However, the definition I have provided comes straight from the beautiful blue books we have all grown to love, the QS manuals. In my opinion, there is very little “wiggle room” on its interpretation.

Section I.2.2.9.2 of the PPAP Third Edition clears up this whole Cpk – Ppk thing.

Click to expand...

thanks for your comment.your 67 words helped me out with the confusion.
and ,totally agree with you about " there appears to be some level of confusion. "

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

Don said: "Going back to basic stats for a moment, the calculation for standard deviation has either one of two possibilities in the denominator: n-1 or n. In school, when you knew the data for the population, you used n. When you did not know the data for the population, you used n-1. The two different types of standard deviation were designated by either sigma (population) or a lower case s (sample). Thus, just using simple logic, you would assume that Cpk is calculated for the population.

I asked the question here once what the AIAG used (I stay as far away from their stuff as I can. Gives me headaches), as the symbol in their method of calculating Ppk and the answer was the lower case s. Thus, it would seem to indicate that Ppk is the process capability for a sample and Cpk is the process capability for the population."

I agree AIAG gives me a headache. But worse are the many consultants and bloggers that are still arguing about WHICH is the LONG TERM CAPABILITY vs the SHORT TERM. Those that use X-bar-R charting for SPC, with suspicious subgroups, are trapped into using within-subgroup sigma in their capability studies....and average that over 100 or more subgroups even though the major source of variation is group to group, not within group! Can you explain how the true variance components impact short vs long term capability metrics?

Re: Ppk vs Cpk - A Good, clear explanation and How Mini-Tab Handles Certain Statistic

well the hacks, bloggers, shamans and AIAG drive me crazy too. The definitions are very well documented and really pretty simple, but everyone has to add their own creative 'twist'. I don't know why. (and whether or not you use the n or n-1 adjustment isn't part of the definition by the way; most people casually use the greek symbol for the sample standard deviation - and it's always a sample standard deviation when calculating either Cpk or Ppk regardless of how those indices are defined -simply because using s is less cool or less confusing as it's universally recognized as meaning some kind of standard deviation).