Process Design FMEAs: What are the best practices?

[ChrisK]

FMEAs: what are the best practices?

We all know FMEA has the potential to be the “front-and-centre” tool for process design: in principle, enabling a cross-functional team to truly analyse and design their process for the preclusion of failure. (My context is within the QS-Big 3 – automotive part manufacturing)

My frustration in real practice has to do with making the FMEA a truly living, alive (excuse redundancy) and central document in the full life-cycle from part and process design through the life of the part.

Unfortunately, the noble aspirations inherent in the writing of FMEAs are often not achieved (do any of these problems sound familiar):

(a) The cross-functional team can’t seem to get together for regular meetings to write good FMEAs.
(b) Although everyone knows FMEAs should be updated after something goes wrong (customer rejection, etc.) No one gets around to updating and revising FMEAs on a regular basis,
(c) The FMEA starts getting thought of as “just another nuisance” document: Sometimes only one individual in the organisation ends up quickly writing it, sometimes after the process has been designed, or worse, they cut and paste from a similar part FMEA in order to have it ready for PPAP and QS audits.

I would be interested in hearing from anyone with success stories in overcoming some of the problems above and/or have implemented a simple, living approach to getting the most out of this powerful tool.

In the meantime, I will mine the wisdom found in this forum for some helpful insights.

All comments welcome. Thanks

 

[Bill Ryan - 2007]

Hi Chris,

You hit my company right on the button. I'm afraid the only success story I can share concerns b).

Because I got sick and tired of answering audit (internal & external) nonconformances, I asked that a note be added to our 8D worksheets which would not allow closure of a quality concern until the appropriate engineer was notified of the corrective action(s) and the PFMEA and CP were updated and issued.

This seems to work fairly well except that the "appropriate engineer" turned out to be me.

 

[ChrisK]

Thanks for your reply Bill: I have done precisely the same thing for the same reason with the same outcome. Paperwork within the framework of a quality management system bogs everyone down...and guess who always fills in the missing gaps!

Any comments from others?

 

[noboxwine]

Hey Chris

I have used the same technique of triggering a FMEA update from the bottom of a C/A. Worked very well and was of benefit on more than one occasion for us. Took a lot of discipline, but doesn’t anything good ? Next, how do you get everyone from the entire cross functional team to set aside four or eight uninterrupted hours to do this quirky exercise for all new parts ? You don't. But, what worked well for me is to prove the FMEA a useful and cost saving exercise (I am the ultimate advocate of the FMEA—one of few AIAG formats I like) to the GM. Then, he/she can mandate an off site meeting to do what I dubbed a “deep dive” FMEA on a product “family” (all steering knuckles, all lower control arm, etc). Pagers and cell phones off. Bring in lunch. Have prints hanging all over the walls, overhead projectors, laptops, parts, and any visual aids that will help. Very soon into the process, all of us played nicely together and uncovered process improvement opportunities never uncovered when the QE does the FMEA in his chair at home the night before PPAP is due. Then you can cut and paste and modify for future PPAP’s, but have already captured the value added for your improving your process. It took a little bit of sales, a large commitment from the GM, but what a return on the few days’ investment. After we became good at it, we began inviting our friends from Gen Motors to be a part of it and they were floored. They requested that they were a part of any future projects. Soon after we began bringing in all suppliers, equipment mfgs and the like. To quantify the return on investment: Single digit PPM across the board. Best of luck !

 

[Bill Ryan - 2007]

noboxwine,

I, too, advocate the FMEA and seem to be harping on them every day. So far, no luck in putting them in the forefront here. I have made a couple of inroads but only due to the software we use which takes Recommended Actions and automatically "blows them across" all parts which may have the same Failure Cause at the same Processing Step. We have saved "stubbing our toes" more than once as the amnesia can't set in because of the documentation.

I like the approach you took and am going to keep on pushing. (By the way - What's wrong with box wine?)

 

[ChrisK]

Thanks noboxwine, and thanks again, Bill for your remarks.

It seems that if people have the attitude that they have to simply “do the FMEA”, then it becomes the “nuisance” document the QE does from home the night before PPAP.

If people would look at the task as “designing a process so that a conforming part is made every time, and do what it takes to make sure it happens”, then the FMEA is probably the tool to get you there. Like all tools, it is only as good as the users will make it. And I know I am stating the obvious. Thanks again and keep the thread going…

 

[noboxwine]

Bill Ryan said:



(By the way - What's wrong with box wine?)

Being a former, single, wine snob, with no responsibilities and too much $, I appauled the notion and image of wine by the litre. Now since I have recently met responsibility, adulthood and planning for the future, this box stuff ain't so bad after all ! Enjoy !

 

[ChrisK]

Thanks to all who posted...is this thread now dead? I would still be interested in hearing successful FMEA implementation stories. Hopefully yours....

 

[Randy Stewart]

What's the intent?

All good inputs to real problems. I would like to add Chris that a CFT doesn't have to be present at one time. I have found that by getting with the department representatives, maybe 2 at a time, works well. Smaller team which is easier to handle and keep in focus and it still meets the intent of a CFT or multidiscipline approach.

 

[ChrisK]

Very good suggestion, Randy. Managing meetings with more than, say, 4 people, unless the meeting is run by a dictator (which is counter to the "spirit" of a CFT, right?) usually don't accomplish much. Using a few people at a time sounds like a useful approach. Chris