How to develop a Control Plan from a PFMEA (Process FMEA)

D

David DeLong

#11
I have always found it interesting developing a Control Plan directly from the PFMEA. It does certainly make sense developing a Control Plan from the Process Flow Diagram but directly from the PFMEA?

When one is involved in a small team developing a Process FMEA, there is a complete section called "Current Process Controls Detection" or "Current Process Controls Preventive". I wonder where the team would obtain the information so that these columns could be completed. Maybe a Control Plan? If there was no control plan developed or even roughly developed, would not the Process FMEA die right there?

Please let me know where one would obtain the information on a new product so that we could complete the controls section of the Process FMEA without a somewhat developed Control Plan?
 

Stijloor

Staff member
Super Moderator
#12
I have always found it interesting developing a Control Plan directly from the PFMEA. It does certainly make sense developing a Control Plan from the Process Flow Diagram but directly from the PFMEA?

When one is involved in a small team developing a Process FMEA, there is a complete section called "Current Process Controls Detection" or "Current Process Controls Preventive". I wonder where the team would obtain the information so that these columns could be completed. Maybe a Control Plan? If there was no control plan developed or even roughly developed, would not the Process FMEA die right there?

Please let me know where one would obtain the information on a new product so that we could complete the controls section of the Process FMEA without a somewhat developed Control Plan?
Hello David,

Even though you may be working on a "new" product or "new" process, it may have a number of characteristics that are pretty similar to a product that has been designed/manufactured before. As you know, the DFMEA and PFMEA process is an assessment of the "current state of affairs", in other words: what controls do we currently have that are either of a preventative or detection type that could be utilized? Historical records combined with "engineering judgment" is the key here. And if there's absolutely nothing in place, then we leave the columns blank which means we've got work to do.

Stijloor.
 
D

David DeLong

#13
Hello David,

Even though you may be working on a "new" product or "new" process, it may have a number of characteristics that are pretty similar to a product that has been designed/manufactured before. As you know, the DFMEA and PFMEA process is an assessment of the "current state of affairs", in other words: what controls do we currently have that are either of a preventative or detection type that could be utilized? Historical records combined with "engineering judgment" is the key here. And if there's absolutely nothing in place, then we leave the columns blank which means we've got work to do.

Stijloor.
Thanks Stijloor

So one would bring Control Plans of a similar products into the meeting and sort of develop it during the Process FMEA. I certainly would not want to be the person representing the Quality group without some sort of preparation.

I have always suggested having a "rough" or "not set in concrete" Control Plan prior to the actual Process FMEA. One should be ready for the meeting and certainly could develop a Control Plan from the Process Flow Diagram prior to the meeting. There may be apparent weakness reflected during the meeting in the prepared Control Plan and, of course, it should be made more robust.

Coming into a meeting without a plan seems unprepared.
 

Stijloor

Staff member
Super Moderator
#14
Thanks Stijloor

So one would bring Control Plans of a similar products into the meeting and sort of develop it during the Process FMEA. I certainly would not want to be the person representing the Quality group without some sort of preparation.

I have always suggested having a "rough" or "not set in concrete" Control Plan prior to the actual Process FMEA. One should be ready for the meeting and certainly could develop a Control Plan from the Process Flow Diagram prior to the meeting. There may be apparent weakness reflected during the meeting in the prepared Control Plan and, of course, it should be made more robust.

Coming into a meeting without a plan seems unprepared.
David,

You're absolutely correct. Obviously, the expectation is that FMEA Team Members would come to the meeting prepared. It also depends on how formal/structured the FMEA Team conducts its business. If the Team has access to the records database on the company's network, they could explore this source of information. Page 23 (Current Design Controls) and page 49 (Current Process Controls) in the AIAG FMEA Manual suggest possible methods.

Stijloor.
 
D

David DeLong

#15
David,

You're absolutely correct. Obviously, the expectation is that FMEA Team Members would come to the meeting prepared. It also depends on how formal/structured the FMEA Team conducts its business. If the Team has access to the records database on the company's network, they could explore this source of information. Page 23 (Current Design Controls) and page 49 (Current Process Controls) in the AIAG FMEA Manual suggest possible methods.

Stijloor.
Page 49 of the AIAG manual only describes both the detection and preventive process controls to consider and does not get into how they are derived.

I guess the point that I am trying to make is this.

Going into a Process FMEA from a Quality department and not having any kind of Control Plan developed is not, what I would consider, being prepared. I am pretty sure that the Process Engineer would not come into the meeting without planning the process either.

The AIAG standard does not state that one should come into the meeting without a Control Plan. It does not state that a Control Plan should be developed through the Process FMEA (at least I can't find it). If one developed some type of Control Plan for new the part, would it be considered reflecting "current process controls"? Again, the AIAG standard is fuzzy.
 

Stijloor

Staff member
Super Moderator
#16
Page 49 of the AIAG manual only describes both the detection and preventive process controls to consider and does not get into how they are derived.

I guess the point that I am trying to make is this.

Going into a Process FMEA from a Quality department and not having any kind of Control Plan developed is not, what I would consider, being prepared. I am pretty sure that the Process Engineer would not come into the meeting without planning the process either.

The AIAG standard does not state that one should come into the meeting without a Control Plan. It does not state that a Control Plan should be developed through the Process FMEA (at least I can't find it). If one developed some type of Control Plan for new the part, would it be considered reflecting "current process controls"? Again, the AIAG standard is fuzzy.
David,

The "normal" flow of events is that the control plan is developed as a result of the FMEA process as well as other assessment activities. "Old" control plans related to similar products/processes can certainly be used during this process.

I am curious about our other Fellow Covers' views. Interesting topic.

David, you mentioned "AIAG Standards." Please note that AIAG manuals are all guidelines and not standards, with the exception of PPAP, which contains "shalls."

Stijloor.
 

Jim Wynne

Super Moderator
#17
I have always found it interesting developing a Control Plan directly from the PFMEA. It does certainly make sense developing a Control Plan from the Process Flow Diagram but directly from the PFMEA?

When one is involved in a small team developing a Process FMEA, there is a complete section called "Current Process Controls Detection" or "Current Process Controls Preventive". I wonder where the team would obtain the information so that these columns could be completed. Maybe a Control Plan? If there was no control plan developed or even roughly developed, would not the Process FMEA die right there?

Please let me know where one would obtain the information on a new product so that we could complete the controls section of the Process FMEA without a somewhat developed Control Plan?
Just because a product is new doesn't necessarily mean that the entire process for producing it is new. If there are no current controls, you just enter "none" and move on. The idea that a control plan is PFMEA output--or part of it--is elementary. Developing effective process controls is the reason for the PFMEA.
 
D

David DeLong

#18
The idea that a control plan is PFMEA output--or part of it--is elementary.
Jim:

Please let me know where in the AIAG reference manual does it states that the Control Plan is an output? Changes are but not the entire Control Plan. If I am wrong here, I will eat crow but I can't find anything the 3rd edition or 2nd edition stating this.

I know that we will review the controls and the potential causes for failure and possibly have changes to both but we do not develop the process during the process FMEA nor nor do we develop the entire Control Plan. Changes to either or both, yes, but developing both during the Process FMEA - mmmmm.
 

Stijloor

Staff member
Super Moderator
#19
Jim:

Please let me know where in the AIAG reference manual does it states that the Control Plan is an output? Changes are but not the entire Control Plan. If I am wrong here, I will eat crow but I can't find anything the 3rd edition or 2nd edition stating this.

I know that we will review the controls and the potential causes for failure and possibly have changes to both but we do not develop the process during the process FMEA nor nor do we develop the entire Control Plan. Changes to either or both, yes, but developing both during the Process FMEA - mmmmm.
David,

From the AIAG APQP Manual, "Control Plan Methodology", Page 35:

For process control and improvement to be effective, a basic understanding of the process must be obtained. A multi-disciplined team is established to develop the Control Plan by utilizing all the available information to gain a better understanding of the process, such as:
  • Process Flow Diagram
  • System/Design/Process Failure Mode and Effects Analysis
  • Special Characteristics
  • Lessons Learned from Similar Parts
  • Team's Knowledge of the Process
  • Design Reviews
  • Optimization methods (e.g., QFD, DOE, etc.)
As you can see, and we have discussed, the Control Plan is the output of the process utilizing all or some the above.

Hope this clarifies.

Stijloor.
 

Jim Wynne

Super Moderator
#20
Please let me know where in the AIAG reference manual does it states that the Control Plan is an output? Changes are but not the entire Control Plan. If I am wrong here, I will eat crow but I can't find anything the 3rd edition or 2nd edition stating this.
Dave, you can use the AIAG manuals as Sacred Received Knowledge if you want to. If you do it consistently, though, you'll wind up dazed and confused. One of the primary reasons that the US auto industry is headed inexorably down the tube is that they have no clue as to how to deal with suppliers, and believe that containers are equal to the things contained, The entirety of the FMEA manual could be condensed to eight or ten pages (without losing anything meaningful) and incorporated into the APQP manual, where it belongs. That'll never happen though, because it makes sense.

There are APQP processes that can (and should) happen concurrently rather than serially. The idea behind the PCP being output of the PFMEA process doesn't necessarily mean that there is no control plan prior to initiation of the PFMEA process. It means that when the triad (PFD/PFMEA/PCP) is complete, there is a logical progression in that order.

For each operation shown on the PFD there is a line item on the PFMEA. During the PFMEA process, each line item is examined for improvement opportunities. The control plans of similar processes may be (and should be) used so as to (A) avoid unnecessary duplication of effort and (B) determine whether improvements to current controls are feasible. If there are no current controls (no similar process or operations) then you simple enter "none" in the Current Process Controls field and move on. When the PFMEA process is complete, a current control plan might be co-opted and modified, or a new one might be created. Whichever happens, the new current control plan will have happened as a result of PFMEA output.

Note also that PFMEA is a process and not a document, and any corroborative consideration for improvement that results in a new or modified control plan may logically (if not strictly literally) be said to result from PFMEA.

I know that we will review the controls and the potential causes for failure and possibly have changes to both but we do not develop the process during the process FMEA nor nor do we develop the entire Control Plan. Changes to either or both, yes, but developing both during the Process FMEA - mmmmm.
I don't think anyone here is suggesting that all process knowledge--documented or otherwise--emanates from the PFMEA process.
 


Top Bottom