Must Key Product/Process Characteristics (KPC's) always be variables?

Phil Huber

Inactive Registered Visitor
Must Key Product/Process Characteristics (KPC's) always be variable? Can they be an attribute go-no-go? Big debate at my company with 4 locations. I say KPC's require monitoring to verify the output of the process is as intended, otherwise reaction must be taken. Comments? Juran or others have definitions? Thanking all of the Cove in advance.
 

Phil Huber

Inactive Registered Visitor
Carol,
YES, my point exactly! :agree1:

In fact, the example you used is one of the points being resisted. I was beginning to think I was wrong in my assumtion.
Thank you again,
 
T

True Position

Technically yes, they should be variable gages but I have had customers agree to attribute gages when the quotes on the variable gages came back.

This mostly comes up with threads and splines in my experience.
 

CarolX

Trusted Information Resource
but I have had customers agree to attribute gages when the quotes on the variable gages came back.

I had a situation like this when a customer wanted a specific test done on pre-production parts. The cost of the test equipment exceeded the cost of the whole original purchase order for the parts. The customer opted out of running the specific test.

Phil - you may want to get a quote on some variable measuring equipment and present it to your group. That may cool some fires.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
Must Key Product/Process Characteristics (KPC's) always be variable?

No. For example, it may be key that during assembly a part is not put on upside down. There is no variable criteria, but it is key - maybe even critical or safety. Could very well be that you will need to verify 100% (poke-a-yoke, vision system, etc.)
 

Bev D

Heretical Statistician
Leader
Super Moderator
they are called key characteristics. they are not called key variables data characteristics. Characteristics can be continuous or categorical and some continuous characteristics are best measured by categorical guages.

This debate seems pretty silly to me...
 

Jim Wynne

Leader
Admin
they are called key characteristics. they are not called key variables data characteristics. Characteristics can be continuous or categorical and some continuous characteristics are best measured by categorical guages.

This debate seems pretty silly to me...

If this is an automotive situation, there might be a need for a variables capability study for PPAP. The 4th Edition AIAG PPAP manual says, in Note 5 on page 7:
For those characteristics that can be studied using X-bar and R charts, a short-term study should be based on a minimum of 25 subgroups containing at least 100 readings...from the significant production run.
I've seen cases where a variables study was done for PPAP and then a switch was made to attributes with the customer's permission.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
If this is an automotive situation, there might be a need for a variables capability study for PPAP. The 4th Edition AIAG PPAP manual says, in Note 5 on page 7...

AIAG starts off in Note 2, page 7 stating that initial process study is focused on variables not attribute data. It does not limit attribute characteristics as being key characteristics, it just claims that they are not applicable to short term capability studies. Now, customers may twist that to infer that they must be variable, but that can be shown to be senseless, as in my previous example. They claim it takes more time to collect data for attribute characteristics, so attribute data is not acceptable for PPAP submission - unless approved by customer. Of course nearly every section of the Initial Process Studies 2.2.11 awash in disclaimers that you "need to get customer agreement", because - quite frankly - even though they want to see capability indices, there are so many exceptions (such as 2.2.11.5) that AIAG does not want to commit to anything.
 

Jim Wynne

Leader
Admin
AIAG starts off in Note 2, page 7 stating that initial process study is focused on variables not attribute data. It does not limit attribute characteristics as being key characteristics, it just claims that they are not applicable to short term capability studies. Now, customers may twist that to infer that they must be variable, but that can be shown to be senseless, as in my previous example. They claim it takes more time to collect data for attribute characteristics, so attribute data is not acceptable for PPAP submission - unless approved by customer. Of course nearly every section of the Initial Process Studies 2.2.11 awash in disclaimers that you "need to get customer agreement", because - quite frankly - even though they want to see capability indices, there are so many exceptions (such as 2.2.11.5) that AIAG does not want to commit to anything.
I agree with all of that, but it usually comes down to what customers expect to see, and not necessarily what AIAG intended or commits to. The common bond between automotive customers is unreasonable requirements.
 
Top Bottom