Error Proofing and integrated manufacturing equipment controls

[Steve.Davidson]

I am in the process of creating a global error proofing process for our 5 semiconductor manufacturing sites, which as you can imagine is not easy, as they all have differing opinions on what constitutes an EP device etc.

Quality Managers for the sites are trying to convince me that IATF requirements for EP does not apply to our industry, as we make billions of products annually, which in my opinion is ridiculous and completely wrong because you cannot just decide to ignore an IATF clause for your own convenience !!.

The processes are many and extremely complex, so my main motivation is to create an EP methodology that works for us and keeps the control plans to a sensible and manageable level.
We have had NC`s raised by our CB related to what they interpret as "EP devices" such as optical inspection sorting equipment , that is integrated into our manufacturing equipment .
My intention and methodology is to exclude these(unless we have added them, as an extra process control for EP purposes) as the list and resulting control plans could be endless , especially when I think about all of the integrated equipment that ensures we manufacture conforming product!.
I will of course include any device outside of the manufacturing equipment that prevents human error, that results in non conforming product.
Just to add, that I have read and understand AIAG CQI-18 Effective Error Proofing and luckily we don`t have any customer CSR`s that require us to be compliant to it as the resources to implement it, would be enormous!.
I would really appreciate your thoughts and perspective on this topic!

 

[Golfman25]

If by error proofing, you mean address potential for human errors, then I would focus on where the human interacts with your manufacturing process, which I assume is highly automated.

 

[Steve.Davidson]

Actually this is the problem, we have human error covered(by EP devices such as jigs included in the control plan etc.) it`s more about IATF auditors defining manufacturing equipment controls (such as automated defect sorting) as error proofing devices and raising NC`s ,which believe me I challenged and threatened to escalate to IATF.
In the end I came up with a methodology (IATF requirement)that excludes these types of devices and was wandering if anyone else has had similair issues?
Thanks!

 

[Golfman25]

Actually this is the problem, we have human error covered(by EP devices such as jigs included in the control plan etc.) it`s more about IATF auditors defining manufacturing equipment controls (such as automated defect sorting) as error proofing devices and raising NC`s ,which believe me I challenged and threatened to escalate to IATF.
In the end I came up with a methodology (IATF requirement)that excludes these types of devices and was wandering if anyone else has had similair issues?
Thanks!

Technically that is an error proofing device vs. human defect sorting. It is what I would put in place to ensure zero defects. A shop down the street from me does a lot of automotive work. Every mold they sell comes with some type of automated detection device. So the operator would remove the part from the mold, and put it into the device for "inspection." Other than loading/unloading, no humans where involved.

 

[Johnnymo62]

By definition, a tester for bad and good is not an error proofing device. Here is IATF 16949's definition.

Error Proofing: Product and manufacturing process design and development to prevent manufacture of nonconforming products.

Error proofing doesn't allow bad parts to be made.

 

[Ashland78]

I work at a big 3 auto company and all of our manufacturing processes here have EPM (Error Proofing Maps), these are controlled documents. It may be worth considering.

These are included in many of our internal audits. Keep in mind PFMEA helps with this as well, as it is part of the risk analysis.

 

[Golfman25]

By definition, a tester for bad and good is not an error proofing device. Here is IATF 16949's definition.

Error Proofing: Product and manufacturing process design and development to prevent manufacture of nonconforming products.

Error proofing doesn't allow bad parts to be made.

Fair enough, but unrealistic. Must be why the auto shops I am familiar with all use automated inspection.

 

[Bev D]

The definition is not unrealistic at all. That is the classic definition form the Toyota production system poke yoke system. Errors create defects. (This is primarily an attributes, human thing, not a continuous data physics thing) The best poke yoke (from waste, cost, customer standpoint) is to prevent errors which prevents defects. The next best is to make errors unlikely which prevents defects. And the next is detect defects if they are created. And a system that automatically detects defects is better than relying on human detection…it’s a system with different levels and we should employ them as needed. In fact many defects are not created by errors at all but from physics variation of inputs and conditions. These defects should have automated inspection where possible. (And based on risk, both severity and frequency of occurence)

Of course I have no idea what IATF says, but they are known to bastardize many good things..

 

[Steve.Davidson]

I work at a big 3 auto company and all of our manufacturing processes here have EPM (Error Proofing Maps), these are controlled documents. It may be worth considering.

These are included in many of our internal audits. Keep in mind PFMEA helps with this as well, as it is part of the risk analysis.

Thanks very much for your reply, I don`t believe I have seen an error proofing map before even in AIAG CQI-18, would you be so kind as to explain a little more in detail so that i can understand what you mean by mapping?