Why is having up-to-date FMEAs important?

Kronos147

Trusted Information Resource
So I would say you might be best served with a master PFMEA. Then you can update that as things change and improve.
I like this. At our shop, the Quality Supervisor started a master PFMEA, if for no other reason, to help create part specific ones.

If there is a quality escape, several of our customer will ask what changes we make to our PFMEA and/or control plan. Others do not ask, but it comes up when they visit.

If the document really is of no use to the production process owners, then, then it becomes a paperwork exercise.

One of the bigger challenges in the Quality field is being a Quality Cheerleader, and being able to cite the successes that using a PFMEA has brought.
 

ARK117

Registered
Follow the process.

I'd expect your organization to have something that describes the manufacturing process (a process flow, written instructions, etc). Follow that, ensure it matches to what is being done, and make sure the the PFMEA, documentation, and the work being done all match.

The PFMEA should cover all the steps, the risks at each steps, the process elements that address the risks (methods of work, inspection, automation, etc). If stuff is missing or incorrect, then the case to dig deep and re-evaluate emerges.
Our documentation is pretty poor as well so this project may serve to get process flow documentation nailed down also. Thank you for the advice, I will keep it in mind as I progress!
 

ARK117

Registered
Several have asked around this question already but let me put it point blank: what's YOUR rationale for wanting to change? If it's just because you feel they are "stale" then you're probably not going to win over any friends. The suggestion @Johnny Quality made is excellent. To add on to that, what manufacturing data do you have? Are the failures occurring at the expected rate? Is the severity of failure properly reflected? Are there new failure modes that weren't captured? (These can't be guesses, they need data to back them up.) So maybe in addition to looking at a gap analysis, if you don't already have data, see what you can do to start collecting.
I appreciate you taking the time to respond to this thread. I apologize if my frustration with my org bleeds into my response.

what manufacturing data do you have? Are the failures occurring at the expected rate? Is the severity of failure properly reflected? Are there new failure modes that weren't captured?
Great questions. In response to all four, I couldn't tell you. I don't know where that information would be, and as far as I know, no one in the org does.

I am not aware of any quality data that's retained other than scrap, rejections, and some basic KPI like productivity/efficiency. SPC is just not a thing that's done here. The systems for SPC are in place, but not used. Maybe more data is recorded than I know, but I think my ignorance on that front is pretty telling. The FMEA aren't stale, they are atrophied. They are not used AT ALL. Not by QA, not by engineering, not by production, no one touches these things unless its to shove in a PPAP.

To answer your main question:
what's YOUR rationale for wanting to change?
I would like to do as much as I can the right way. Learn what I can. I'm early in my career, and I'm worried I'm developing bad habits because instead of learning how to handle these fundamental QA/engineering tools from experts, I am building some of these tools/systems with minimal guidance. I want the FMEA updated because as I understand it's a basic foundational tool of QA, and I'm frustrated that I am just now learning about it. I don't want the next generation to have to feel the same frustration and ask "why hasn't anyone fixed this?" like I am now.
 

Golfman25

Trusted Information Resource
I appreciate you taking the time to respond to this thread. I apologize if my frustration with my org bleeds into my response.


Great questions. In response to all four, I couldn't tell you. I don't know where that information would be, and as far as I know, no one in the org does.

I am not aware of any quality data that's retained other than scrap, rejections, and some basic KPI like productivity/efficiency. SPC is just not a thing that's done here. The systems for SPC are in place, but not used. Maybe more data is recorded than I know, but I think my ignorance on that front is pretty telling. The FMEA aren't stale, they are atrophied. They are not used AT ALL. Not by QA, not by engineering, not by production, no one touches these things unless its to shove in a PPAP.

To answer your main question:

I would like to do as much as I can the right way. Learn what I can. I'm early in my career, and I'm worried I'm developing bad habits because instead of learning how to handle these fundamental QA/engineering tools from experts, I am building some of these tools/systems with minimal guidance. I want the FMEA updated because as I understand it's a basic foundational tool of QA, and I'm frustrated that I am just now learning about it. I don't want the next generation to have to feel the same frustration and ask "why hasn't anyone fixed this?" like I am now.
Sorry, but I view the FMEA as a planning tool, not really an operating tool. So in our case when we sit down and review a production process, 99.9% it's based on something similar. We identify any of the typical failure modes we may find, and then if it's something new to us, we may add something to our master FMEA. It's a high level document and once your processes are in place, you'll refer to it only when changes are needed.
 

ARK117

Registered
Sorry, but I view the FMEA as a planning tool, not really an operating tool. So in our case when we sit down and review a production process, 99.9% it's based on something similar. We identify any of the typical failure modes we may find, and then if it's something new to us, we may add something to our master FMEA. It's a high level document and once your processes are in place, you'll refer to it only when changes are needed.
When are changes needed, then? Would you not use it in response to rejections or to build your control plan?
 

ARK117

Registered
One of the bigger challenges in the Quality field is being a Quality Cheerleader, and being able to cite the successes that using a PFMEA has brought.
I feel this one in my bones.

I think we use something like what you are referring to. We have several top level PFMEAs for each possible interaction of processes, but not for individual products. Is that kinda what you mean?
 

Golfman25

Trusted Information Resource
When are changes needed, then? Would you not use it in response to rejections or to build your control plan?
As yendor said, it depends. First question is whether the failure mode was identified. If it was, then maybe your corrective action could change some things — occurrence etc. If it wasn’t identified, then a good time to add.
 

toniriazor

Involved In Discussions
I just had a discussion with my plant manager where I argued that we should update our 10+ year old FMEAs. They weren't convinced by my arguments, but they are open to investing the man power into updating them if I can justify it. Any suggestions?

Context: I work in a sheet metal fabrication facility. The whole QA department got cycled within the last 3 years. Professionals that had decades of experience either retired or moved on to different companies, leaving new grads (like myself) and other inexperienced personnel to fill the void. Since I've been here, I have never even looked at the FMEA because it just isn't something we were ever told/taught to use, and that feels like a massive oversite. With that in mind, how big of a deal is it that we don't have updated FMEAs or that we don't use it?

Does your plant manager visit the shop floor? If yes, you could show him how FMEA has helped to assess a risk in production and contain it. Or you could walk with him through some of the failures modes and potential causes that have been identified and corrected.

FMEA is a "living" document and if there is a generic FMEA in your company that it means - maintaining it shouldn't be that complex at all and should not require even a new person. You could negotiate to take it yourself and be given certain benefits, instead of hiring new person. Upon failures you could "review" it and if deemed necessary update it.

I recently started building a PFMEA (self-initiative) in already existing production and it is another story when you are behind it, as an owner. I used to use it a lot in the automotive company I worked for before, but there was another person in charge from engineering team. I don't dedicate much time to it now, as this is something I do voluntary, but it already helped me identify several big gaps in the processes that I required to be addressed immediately by engineering and fixed. FMEAs are time-robbing activity when there is no base at all, but if there is 10+ year old FMEA as you say, I would assume that most of the risks are already there and it shouldn't be big deal to maintain it and add new potential causes and respective actions.

Why is important to be updated? Well this is an automotive quality core tool and one of the most audited processes within the QMS (internal/external) and the actions in FMEA trigger action on the other important tools used to control processes/parts such as Control Plan and Work Instructions in the shop floor.
 
Last edited:

Bev D

Heretical Statistician
Leader
Super Moderator
After reading through the thread I think that the real issue here is not FMEAs, but your knowledge, proficiency and experience with Quality Engineering methods and tools. This is exacerbated by the lack of quality awareness in your organization. There is nothing wrong with the advice you have given so far, yet it is off the mark a bit. You really can’t put an FMEA process in place until you are proficient in FMEAs. And it doesn’t help to put this advanced process in place when the organization doesn’t even value or use the basics like SPC and failure data. I actually went in to my last coupe of organizations that were similar in their deployment of quality methods; I found the best place to start is with the basics: Problem Solving and SPC. Once the organization understands how to solve problems and is collecting the appropriate data to indicate when a problem has occurred, things like FMEA will make much more sense - not only in why they should do it but how to do it…other wise it’s just an exercise in filling out a (long and tedious) form and then moving on. They will only resent you for it.

There are several resources that you can study listed here in “Essential References for Quality Engineering”. Most are free.
 
Top Bottom