AIAG-VDA FMEA tutorial and completely filled FMEA

AndrewK

Involved In Discussions
I think so, too.
Materials, including consumables, should be mention the presence or absence of the Element Function necessary to produce the required of the Process Step Function.
However, in FMEA, assumes that the incoming materials are correct.
Why occur that you say? - Wrong paper type, wrong paper size
Also, isn't "similar paper type allocated nearby" classified as "Environment"?
it depends what you are considering under environment.
I usually put there requierements for lightning, power supply, clearanse, etc.
Similar types of material in my opinion is not belongs to environmental conditions - it is attributes of materail itself (cause might sound just like : Similar material), but as i said before- that things are subjective and depending how we can prooof our conclusion. I haven't met any guide or classificator regarding 4-5 Ms for Isikawa or else- might be if it exists, there iw written some logic how and what to assighn. I rely on My Quality team during analysis, and our common thinking is in most cases like i describe.
 

Bev D

Heretical Statistician
Leader
Super Moderator
Bev D,
Don't make me say the same thing over and over again.
As you can see from what I've written above so far, the real examples contain confidential information, so I've replaced them with simple examples.
Then, after fully understanding the importance of the existing 4M (Material, Machine, Man, Environment), we are discussing the case where "Method" must be added to this. We are not ignoring the existing 4M.
If you also don't know how to write "Method", figure it out together. Your points are all inept and do nothing for how to write about "Method".
Even though the AIAG VDA FMEA handbook says that "Method" can be added according to the company, we are confused because the handbook does not provide specific examples of how to write it.

AndrewK,
I'm not native English speaker, too.
So maybe I didn't fully understand what you were trying to say. I'm sorry for taking too long to understand.
What you're saying is, in Structure Analysis, it's better to write something like a proper noun instead of writing a specific recipe or condition? - as shown in the table below.
STRUCTURE ANALYSIS (Step 2)<-FUNCTION ANALYSIS (Step 3)<-FAILURE ANALYSIS (Step 4)
3. Process Work Element 4M Type<-3. Function of the Process Work Element and Process Characteristic<-3. Failure Cause (FC) of the Next Lower of Characteristic
MethodManual steering methodMethodEstablish proper mixing speed according to work instruction #12345The correct information is not conveyed to the operator
Failure Cause; Indeed, Failure cause is opposite of requirement of process element.
In other words, stiring at a different mixing speed than work instruction #12345 is FC in [01] Mixing of powder.
At this time, if the mixing speed was different due to human error by the operator, it should be categorized as Man element.
So, we thought it was Method's FC that the correct information is not conveyed to the operator.
Isn't there a contradiction in this way of thinking?
Simple questions often have complex answers. FMEA is a complex quality and engineering ‘tool’. It requires deep knowledge and purposeful thought to get it right. So answers to even simple questions can be complex.



To reiterate (for others who will read this thread): the 6Ms are a thought device to stimulate deeper thought in order to thoroughly analyze your process. It is not required to categorize things as one of the Ms nor is it required to have something for every M. Another thing to remember is that the Ms are not necessarily mutually exclusive.
So let’s discuss “Method”…the most helpful thought tool is probably a simple flow chart that shows what is done in sequence and words that describe how it is done. In a PFMEA the ‘method’ will include what and how. Remember too that a real PFMEA (as opposed to an after the fact, fill out the form to check the box approach) is an iterative process using scientific knowledge and experimentation to improve the process.
The Design will specify the ingredients and quantities. It will also specify material properties. The Process will specify what is done, how it is done and in what sequence as well as operating parameters and equipment. A work instruction is the embodiment of the process method.

In the simple example of mixing to create pancake batter we need to understand the method of mixing. The method is much more than simply how the ‘batter’ is stirred. As any baker knows there is a sequence to mixing that is critical. Also the type of ‘mixing device’, speed and duration are critical. The recipe - or design - for pancake batter usually includes some dry ingredients (flour, sugar, salt, baking powder…) and some wet ingredients (eggs, milk, melted butter…). If the function we are assessing is mixing and failure mode we are concerned with is ‘lumpy batter’, then we need to consider several things:

  • Is the flour sifted or simply measured as is? What is used to mix the flour? Flour has a moderate Carr index which means that it ‘compacts’ easily and this causes lumps of compacted flour. The compaction occurs when air is released from the flour. This happens during natural settling form flour that sits in a package and the package is moved around. It also happens if the flour is over mixed.
  • Are the dry ingredients and wet ingredients all mixed together at once. HUGE lumps will occur. You also should never add melted butter by itself into dry ingredients. Established scientific knowledge informs us that the dry ingredients and wet ingredients should be mixed separately and then combined with a gentle folding (using a spatula or spoon, not a whisk). The dry ingredients should be stirred together lightly so as not to remove air. The eggs should be whisked first (have you ever tried to chase an egg around liquids? Not fun) then the other wet ingredients are mixed in using a whisk to break up the viscosity of the eggs and incorporate all the wet ingredients evenly or excessive lumps will occur.
  • Another aspect is how the ingredients will be measured. (Here comes the Measurement M in conjunction with the Method M) Some recipes use weights and others use volume. Do the measuring devices match the units of measure used for the ingredients in the design? How is the flour measured: is it measured in a dry cup and leveled off (easy for an operator to do) or is it poured into a liquid measure it cup and ‘eyeballed’ by the operator? Of course this loops back to whether or not the flour is sifted, as sifted flour has a lighter weight (less flour, more air) than the same volume as un-sifted flour…
I could continue about Method but I think everyone will get the point. We must go deeper than just referencing an instruction and the operator’s ability or competence to follow it.
Remember that filling out the form should take the least amount of effort and time.

I always suspect any FMEA that points to the instructions without detailing the instructions. In my 40+ years of practicing, teaching, coaching and speaking about FMEA I have found that this signals a shallow approach to simply blame the operator for anything that goes wrong. In practice it is rarely the operator that is at fault. It is far more frequent that the designated method itself is flawed…it either allows way too much variation or it is simply incapable of producing consistent results.
 

Bev D

Heretical Statistician
Leader
Super Moderator
Traditionally, a PFMEA assumes that the previous steps (and the design) are correct and thus the correct materials are assumed to be delivered. (Otherwise we are double-dipping functions and failure modes). However it is possible that the wrong material is selected by the operator (Man) even tho the correct material is present. There are many causes and solutions to this. Going back to the pancake batter, sugar and salt look alike (and are often confused by novice bakers). If the sugar and salt are in similar sized, shaped and colored packages and the materials are not laid out in a specific sequence (method adn environment) it is possible and even likely that the operator may mistake one for the other. Environmental conditions like lighting and pressure to produce can exacerbate this tendency…Man often makes errors when the method and environment enable errors…

Remember that it isn’t about assigning a M, it’s about using the Ms to guide your thoughts and analysis of the process.
 

Sleepy Cat

Starting to get Involved
Bev D;
> It is far more frequent that the designated method itself is flawed.
We strongly think so too. So we thought to add Method to our P-FMEA (just a several steps).
It's also frustrating that the handbook doesn't mention Method, even though it's pretty important.

> it’s about using the Ms to guide your thoughts and analysis of the process.
We strongly agree with this as well. The important thing is the risk management for critical items from the analysis, not the assignment of M.

However, on the other hand, we have to meet the customer's requirement to fill the format in which we have to write every M element.
Even if FMEA says that each company has its own way of writing, we don't want to write too much irrelevant things, and we would like to create a file that is useful for risk assessment, even if only a little.


Bev D and AndrewK's suggestions made me realize my mistake.
I forgot that Failure Cause in P-FMEA refers to variation in the Function of Process Work Element. - that is, variations that affect the final product.
The word "Cause" confuses me.
Each Failure Cause is a factor that may occur a defect, and is not exactly a "cause", right? - for example, the “root cause” obtained by 5-whys analysis.

Bev D;
Thanks for the detailed explanation about Method and others. I tried to show what you explained in Fig. 1, is this correct?
This is probably the content that leads to Function Analysis in Step 3.
On the other hand, Fig.2 shows a diagram of applying Step 4 with FTA-like thinking.
Of course, I am fully aware that I should write "Problems" or "Effects" on the head of the fish in the original Ishikawa diagram (fishbone diagram). However, I thought it would be helpful for thinking about P-FMEA, so I dared to draw it like this. Please forgive this.
It is the same as Fig.1 that the first bone is divided into 6Ms. On the second bone, I wrote some important items that Bev D gave me.
However, in Fig.2, there are two things written on the last third bone. For example, how to stir - remove air has two sentences: "how to stir" and "remove air".

AIAG-VDA FMEA tutorial and completely filled FMEA


I can add a smaller bone (4th bone), but that will not match the P-FMEA column. Therefore, I drew Fig. 3 that the bones according to the column of Step 4 of P-FMEA.
This is no contradiction with Fig.4, which is drew Step3 in the same way.
(Please note that the top and bottom order is reversed from Fig. 1 & 2.)

AIAG-VDA FMEA tutorial and completely filled FMEA

Applying the Material example given by AndrewK to Figures 3 and 4 results in Figures 5 and 6.
There seems to be no contradiction in this combination too.
AIAG-VDA FMEA tutorial and completely filled FMEA


By the way, applying the example given by Bev D (same content as in Fig. 2) to Fig. 3 results in Fig. 7.
AIAG-VDA FMEA tutorial and completely filled FMEA


Can we consider this to be the classification method for Step 4 in P-FMEA?

And the following things that Bev D taught us will be connected horizontally when we do a 5-whys analysis, as shown in Figure 8.
> the Ms are not necessarily mutually exclusive.
> If the sugar and salt are in similar sized, shaped and colored packages and the materials are not laid out in a specific sequence (method adn environment) it is possible and even likely that the operator may mistake one for the other.
> Environmental conditions like lighting and pressure to produce can exacerbate this tendency…Man often makes errors when the method and environment enable errors…
AIAG-VDA FMEA tutorial and completely filled FMEA
 

Bev D

Heretical Statistician
Leader
Super Moderator
So I cannot really help if VDA or your Customer actually wants you to categorize things into the “6Ms”. Because this is stupid and only adds unnecessary documentation to (what I supposed to be) a very detailed engineering process. I also do not support fishbone diagrams as the 6Ms are not mutually exclusive and and the structure does not allow one M to lead to another M to lead to another M like it happens in real life. Fault Tree Analysis will work for causal analysis but Fishbones are diversionary and confusing as they break teh natural cause and effect chain.

To clarify the “cause” is the factor and factors that create the failure mode. The failure mode is the effect of the cause in terms of how the function fails to execute properly. Factors are inputs to a function and a failure mode is a result of a failed function.
 

Sleepy Cat

Starting to get Involved
Bev D, Thank you for your kind and candid response. But I deplore your answer too much.
How good would it be if we could say to our customers, "It is stupid to have to submit P-FMEA." However, IATF's PPAP requires the submission of P-FMEA, and without it our company will not be able to continue operating.
What is needed now is not academic discussion, but "AIAG-VDA FMEA tutorial and completely filled FMEA" as the title of this thread says.

I may ask you to teach me something for some time in the future, but I would appreciate your help at that time.
 

Bev D

Heretical Statistician
Leader
Super Moderator
Well as I have said many times, I will not help someone do something that I know is diversionary, a waste of time or wrong. I am here to promote good quality methods. I do believe strongly in FMEA but do not enable the use of fishbone diagrams or assigning a cause in an FMEA to one of the 6Ms. I have clearly stated why this is wrong. If your Customer or VDA requires this I feel for you. But my mission is to stop that kind of insanity, not enable it. Perhaps there is someone else here who will help you with the administrative task…
 

Sleepy Cat

Starting to get Involved
Bev D,
I think I understand your point. While we were discussing FMEA, the purpose was different. That's all.

Oh, let me add a little. The AIAG-VDA FMEA Handbook recommends using Ishikawa diagrams (fishbone diagrams) for Failure Analysis, so I followed them.

It is doubtful whether our customers will be able to correctly judge whether the FMEA that we have worked so hard to write is correct or not...:confused:
We will do our best to write it up.
 

Bev D

Heretical Statistician
Leader
Super Moderator
It is doubtful whether our customers will be able to correctly judge whether the FMEA that we have worked so hard to write is correct or not...:confused:
We will do our best to write it up.
You are correct. They probably won’t even read it. Just look at it to see that you did something. At least that has been my experience. Even more reason to focus on the FMEA as a way to help you make a better product.
 

AndrewK

Involved In Discussions
Agree also! More over usually(at least in auutomotive - from my knovledge) FMEA is done for internal purposes to develop robust product or process - not for the customer. In handbook also mentioned that FMEA analysis is highly subjective and represents teams expirience and knowledge and cannot be compared with similar analysis from other teams. So.. correctness or incorrectness of analysis may be only if it comply to basic methodology rules or not ( Scoring rules basically, rarely structure analysis)
 
Last edited:
Top Bottom