I am trying to understand the difference between Cpk and Ppk and when to use them. I think I have it sussed, but I am not sure. I have been trawling through reams and reams of literature until my head spins. OK. So far, I think I should be using Cpk when the process is in control and the data is normally distributed, and Ppk when the process is unstable and/or the data is not normally distributed. Are my assumptions right? I am also confused about the terms 'Short term' for Cpk and 'Long Term' for Ppk. I have read on this forum that, "Cpk attempts to answer the question, does my process in the long run meet specification?. So why is Cpk short term? Any help and guidance appreciated.

David

All that you said is right, but going to the bones Cpk uses within_sample_variation and ppk uses total_variation as estimate of standard deviation.

The "long term" and "short term" terminology, I think became in use, when at the start of production of a new product on the line, the first estimates of indicators are labeled "long term" (MIL-STD-1916), taking in account all the variation of the process (ppk). Also, you may spect little variation in a smaller time period, an so more stability (cpk).

When to use them?, as you said

CPk when the process is in control and the data is normally distributed, and PPk when the process is unstable and/or the data is not normally distributed


Many companies ask the providers for a cpk, so it's a difficult point to say "no, I don't have stability, could I give you ppk insteed?".

And for you?, being practical, both indicators are good, but keep on mind that if you can calculate Cp or pp with the Cpk/ppk, calculate them, because both indicators tell more than just one, Cpk doesn't substitute Cpk but complement it, for example if you have a Cpk value can be because of the centering, or the variation. And take also in account something else that most forget, all indicators are estimates and have variation asociated with them depending on the sample size.

:thedeal:

Try this link:
More than you want to know about capability! (http://www.qualityadvisor.com/library/capability/capability-menu.htm)

Also (I may get corrected since I don't have a copy yet) try the Automotive Industry Action Group Statistical Process Control Manual. U$30, at *** DEAD LINK REMOVED ***

Short course: C analyzes a system's aptitude to perform, P measures a system's actual process performance.

Hi! everybody,

I´m having some difficulties regarding the following:

Is there any minimum value to apply for sampling events? :confused:

I mean when you perform a study bias, stability,or so and production rate is 10,000 pieces a shift. How many pieces must I study to make it representative?

Is there a minimum or can I specify it? :frust:

Thanks in advance

:bonk:

Ford Motor Company was one of the first to use the term Cpk, defined as simply process capability (determined after a process was in control). For production environments it meant by default "long term" process capability.

Ppk first appeared on initial approval requirements (a.k.a. PPAP) for first run type capability (after control was established). Since the data was pulled from the entire "first run" population it was called Ppk. But it was in reality a short term process capability, or process (P)otential.

When regular production started (i.e. unlimited population), the statistical studies could refer to Cpk, or process (C)apability. This is long term capability.

However, during the decades to follow, the quality community began to focus on the source of the sample only, and the short/long Cpk/Ppk terms got reversed. As Icy Mtn states, "C" is aptitude to perform, "P" is actual performance. You can only be convinced of actual performance in the short term. With ongoing production, all you'll ever know is "aptitude to perform".

NOTE: They both are calculated the same, and both assume a controlled process.

Welcome to the Cove, Rob. Very good first post.

I think (though not quite sure), Minitab mixed up the long-short terminology. I understand they have now dropped it in favor of within & overall.

When using these terms, one must be aware of the method & period of data collection, state of the process, the way sigma is calculated...etc.

I'm new to the forum, and thought I'd just pass on my limited thoughts about Cpk. In my mind, I look very analytically at the definition which I interpret as:

How many 3*(standard deviations) the mean is from the closest specification limit.

From there if you know anything about the shape of the distribution or how the "standard deviation" was calculated you can start understanding in more detail what your process might deliver.

And I always like to remember a favorite quote:

"All models are wrong, some are just more useful than others"



Jeff

Hi Jeff,
Welcome to the Cove.

I will look forward to your posts and if you ever need any help or advice in the quality field, this is the place to come.

Some of us have even been know to "give advice" outside the quality field.

Dave

First, I introduce myself.
My name is Guo Qing.
For Cpk & Ppk, my opinions as follows:
1.Cpk indicate for a stable process. Ppk indicate for a unstable process.
2. The computing method for standard deviation for Cpk & Ppk is different.
You also can study the difference of Cpk & Ppk through their respective formula.

Thanks and best regards
Guo Qing :)

Welcome to the Cove Guo Qing. :bigwave:

You input is appreciated. However, since you cannot determine capability without first controlling your process, both Cpk and Ppk require a stable process.

Welcome to the Cove Guo Qing.

Dave

They are both calculated in the same fashion, but with different estimates of standard deviation.

From the AIAg manual:
Ppk(the performance index)- Uses sigma-s defined as "the estimate of the standard deviation of a process . . . . . ".

Cpk(the capability index) - sigma - R(avg)/d2 defined as "The estimate of the standard deviation of a stable process . . . . .".

Leading one to believe that Ppk can be calculated from an unstable process. It appears to me that Cpk would be the short term and Ppk the long term, however, I like the terms "capability to perform" and "continued performance".

Sigh..... :rolleyes:


Leading one to believe that Ppk can be calculated from an unstable process.

That is quite a stretch of an assumption. The lack of the word stable does not imply otherwise. Pages 13-18 of the AIAG SPC Manual explains it quite clearly. For example, it says "In short: the process must first be brought into statistical control [stabilize the process]... Then its performance is predictable, and its capability... can be assessed".

I am agree with Sam,
The use of Total variation for the standard deviation estimate take in account the changes of the mean or the stability of the variable.

If the process is stable it should not show any "long term" changes, so Cpk equals ppk.

The question here must be "if ppk worth the calculation if is the process is non stable", I tink as I readed somewere "To manage you should measure" and ppk or cpk are measures of the process behabiur. The study of the indicator variation could be of help to understand and show changes on the process.

Do both of these take into consideration the variation within subgroups, or variation between subgroups? I have been confused about which option is proper. Of the data I have reviewed, the better result has been to chose the variation between subgroups. This was a software option to chose for determining Cpk.

Cp/Cpk vs Pp/Ppk (Short term using population sigma) - Formulas to use? (http://elsmar.com/Forums/showthread.php?t=4048)
Capability: Short Term or Long Term? Cpk= Long term and Ppk= Short term? (http://elsmar.com/Forums/showthread.php?t=6444)

and the links (Similar threads) at the bottom of this page

Ralph
Do both of these take into consideration the variation within subgroups, or variation between subgroups? I have been confused about which option is proper. Of the data I have reviewed, the better result has been to chose the variation between subgroups. This was a software option to chose for determining Cpk.

For Cpk is within variation and for Ppk Total.

The between subgroup is not documented as used for any index (as far as i know), but as i found for individuals with too much correlation, sometimes one has to be practical (i tink many will not agree with me, but if it suits you, use it) and Total, within and between variation are just aproximations to the real thing "the standard deviation". The problem is that is well documented that both Total and between subgroup variation are affected too much by outliers.

:truce:

Sigh..... :rolleyes:




That is quite a stretch of an assumption. The lack of the word stable does not imply otherwise. Pages 13-18 of the AIAG SPC Manual explains it quite clearly. For example, it says "In short: the process must first be brought into statistical control [stabilize the process]... Then its performance is predictable, and its capability... can be assessed".

refer to section 5 of the AIAG manual; Ppk = performence = total variation.
I think this may expalin it better.

Do both of these take into consideration the variation within subgroups, or variation between subgroups? I have been confused about which option is proper. Of the data I have reviewed, the better result has been to chose the variation between subgroups. This was a software option to chose for determining Cpk.

Look at sefction 5 B.3 of the AIAg manual. That explaination may help.

Has anyone tried minitab software. It defines cpk as short term capability and ppk as long term capability. The difference simply lies on the formula. Cpk uses the D2 constant and Ppk uses the standard deviation.

Please read this entire thread and do a search of similar threads. Much has been written and debated.

Minitab's definition serves only to confuse the issue. However that stats used in the denominator are correct - Cpk uses an estimate of sigma, generally attained through control charts (which presupposes long term) and Ppk uses sigma (standard deviation) because the entire population is known (suggesting short term), and so the "performance" is known.

But, then again, don't worry about it. Either way, you get a good idea of how well you are staying within the spec tolerances, without being too nit-picky (assuming, as always, your process is in control).

Has anyone tried minitab software. It defines cpk as short term capability and ppk as long term capability. The difference simply lies on the formula. Cpk uses the D2 constant and Ppk uses the standard deviation.
That's almost correct; standard deviation comes in two varieties: Sample and Population. The former is used in calculating Cpk and the latter in Ppk calculations. The Greek lowercase sigma (s) is widely and incorrectly used in quality to represent both varieties, but sigma correctly identifies the population standard deviation while "s" is used to represent the sample standard deviation. Here's some more information:Wikipedia: Standard Deviation (http://en.wikipedia.org/wiki/Standard_deviation)

MINITAB 14's process capability terminology matches that of the new second edition of the AIAG SPC reference manual. MINITAB was ahead of the game there.

There is an odd problem though. In order to have MINITAB 14 match the formulas found in the AIAG SPC manual, you will need to do the following:

Starting with the MINITAB default settings, when using either the Capability Analysis or the Capability Sixpack, go to the Estimate options that effect the estimation of the standard devaition. The ONLY way to match the AIAG SPC manual completely is to:

1. Only use the Rbar method of estimating the standard deviation, and
2. Uncheck the "Use unbiasing constants" checkbox. This is NOT the default.

This will provide the correct within and overall standard deviations.

The within standard deviation based upon Rbar will use the biasing constants regardless choice (an biased Rbar estimate doesn't exist), which is what AIAG uses. The overall standard deviation will be biased. This is what you want. This is the sample standard deviation that you learned in school and the one AIAG uses.

Unfortunately you will have to uncheck the unbiasing constants checkbox everytime you use these tools. There is no preference setting in Options that will take care of this for you without messing up other stuff. I've told Minitab about this and I'm sure they'll correct it soon.

If you select the Sbar method, one of the two standard deviations will be incorrect, depending on whether the ubiasing constants checkbox is checked or not. If it is checked, then the overall standard deviation will not be correct. If the box is NOT checked, then the within standard deviation will not be correct. You can't win using Sbar. Minitab needs to fix this little oddity.

Leave the unbiasing constants option in Tools > Options checked - this is the default - or else many of the control chart calculations will be incorrect.

does it mean that even AIAG will soon release its new edition and we will again have new understanding of the terminologies?

KenK!

thanks for pointing out the problem with minitab.


We have a computerized SPC system here that automatically calculates those statistics, but I couldnt figure out why pasting the same info into Minitab would give me different results!


You Rock!

:biglaugh: :D

Thread 'bumped' in response to some posts in this recent discussion thread: What is the meaning of "k" in Cpk and Ppk? (http://elsmar.com/Forums/showthread.php?t=15164)