Process capability or Process control - Which comes first?

[jrubio]

Tim.

Is not the same Stability than Statical Control.

Stability: Your sigma is changing

Under Statical Control Process: your distribution is known and the formulas are previously calculated: Average, Sigma.

i.e Gauss.

Average=1/2PiSigma*Exp(.....

 

[jrubio]

Tim I disagree

Ppk is the capability when the process is not stable. Please Refer to
PPAP Manual page 7 Third Edition.

When the process becomes stable the index must be Cpk.

Please read page 7

I will translate exactly the sentence, May be I am wrong.

Sorry for My English is getting worse.

"It can be calculated the Cpk index if the process is stable, PPk it should be used for a process Chronic unstable but with a result that meet specification"

 

[jrubio]

I was thinking specifically about statistical process control when I was saying "control". As you note, it can also be used to mean the checks that are in place, as in "document control" or "control of non-conforming products".

Tim F

If you mean statistical process control, I agree but take care that you can control your process without Statical modes.

i.e Oil Temperature less that 40ºC

It does not matter the distribution of points if less than 40 the process is controlled.


You can also control it via Atribute mode.

i.e If the red-cross is located in that point the process is controlled.

 

[jrubio]

I will translate exactly the sentence, May be I am wrong.

Sorry for My English is getting worse.

"It can be calculated the Cpk index if the process is stable, PPk it should be used for a process Chronic unstable but with a result that meet specification"

Example.

1) 5 parts Av=2 Sigma=3 ppk1=2.1
2) 5 parts Av=2 Sigma=2.9 ppk2=2.3
3) 5 parts Av=2 Sigma=2 ppk3= 2


ppk due to the Capability is changing, Not Stable,
...

17) 5 parts Av=2 Sigma= 1 ppk97=2
18) 5 parts Av=2 Sigma= 1 Cp1=2
19) 5 parts Av=2 Sigma= 1.01 Cp2=2.1
20) 5 parts Av=2 Sigma= 1.01 Cp2=2.1

Now Stable finished.

Total = 100 parts.


The only serious doubt I really have is the minimumn number of points needed to say than a distribution is normal, 5, 10 ??????

 

[jrubio]

If you See Delphi Template, The template calculate both indexes which are different.

 

[jrubio]

Then because if the singular samples (Lets say 5) fit a distribution normal, if you add all the points (20 shifts) total 100 points and make the Histogram the points follow also a normal distribution by one Statical Theorema.

That´s why in Delphi templete the 100 samples is map in a Histogram. Obviously is the total parts do not follow the normal distribution some of th shift are not following the normal.

 

[jrubio]

The only serious doubt I really have is the minimumn number of points needed to say than a distribution is normal, 5, 10 ??????[/QUOTE]

I woud say 5.

3 not althugh the Gaus is simetric,
5 in order to see the Kurtosis.

But in fact I can not proof.

 

[jrubio]

Example.

1) 5 parts Av=2 Sigma=3 ppk1=2.1
2) 5 parts Av=2 Sigma=2.9 ppk2=2.3
3) 5 parts Av=2 Sigma=2 ppk3= 2


17) 5 parts Av=2 Sigma= 1 ppk17=2


ppk due to the Capability is changing, Not Stable,

You can calculalate now the Long term Capability PPk for all the 17*5 parts
...


18) 5 parts Av=2 Sigma= 1 Cp1=2
19) 5 parts Av=2 Sigma= 1.01 Cp2=2
20) 5 parts Av=2 Sigma= 1.01 Cp2=2

Now Stable finished. You can calculate the Cpk for all the 3*5=15 parts.

Total = 100 parts.


Also it can be calculated the capability for all the points 100 but it will be worst that in really fact is, But I think no problem due to meet spec.

QUOTE]

 

[Bev D]

Tim I disagree

Ppk is the capability when the process is not stable. Please Refer to
PPAP Manual page 7 Third Edition.

When the process becomes stable the index must be Cpk.

Well, the AIAG methods don't always line up with what the rest of industry does...so the use of Cpk and Ppk for Automotive and for everyone else is different. that cases a LOT of undue confusion.
But, a long long time ago I understood this so my memory might not be correct and I no longer have the AIAG manuals (thank what deity you believe in - or not) but I'll take a shot the Ppk for unstable processes: I THINK I REMEMBER that the comment you refer to is that if you have a process that is in fact in statistical control then it must have a Cpk of 1.33. If your process is NOT in statistical control then it must have a Ppk of 1.5 (or higher, I don't remember the exact numbers).
Please note that I may have remembered this incorrectly adn I certainly don't endorse it. (In past posts I have stated my belif that all capability indices should be eliminated.)

 

[jrubio]

Well, the AIAG methods don't always line up with what the rest of industry does...so the use of Cpk and Ppk for Automotive and for everyone else is different. that cases a LOT of undue confusion.
But, a long long time ago I understood this so my memory might not be correct and I no longer have the AIAG manuals (thank what deity you believe in - or not) but I'll take a shot the Ppk for unstable processes: I THINK I REMEMBER that the comment you refer to is that if you have a process that is in fact in statistical control then it must have a Cpk of 1.33. If your process is NOT in statistical control then it must have a Ppk of 1.5 (or higher, I don't remember the exact numbers).
Please note that I may have remembered this incorrectly adn I certainly don't endorse it. (In past posts I have stated my belif that all capability indices should be eliminated.)

In fact what Customer would Require is the estimated ppms in terms of probability,

Cpk=1.67 I think it was 0.03ppm

It can be deducted the formula

Ppm= P (LT>Normal(Av,Sigma)>UT)/10^6