My first post! I've been hanging around here for years and I have gotten so much help from the forums! I've never posted because a) I've always found the answers in a previous thread, or b) if never felt educated enough to help answer someone's question.

My company has had Process Flow Diagrams, PFMEAs and Control Plans for many years, but they've been neglected until recently. Only last year did customers start complaining that our PFMEA was not in order of the process flow! So I'm currently working on revising our PFMEAs to match the process flow set in the Process Flow Diagram.

My question is this: Should inspections be included as a step in the process flow? Our current Flow Diagrams include 2 inspections as step. One is a visual inspection that occurs on every piece. The other is the final inspection that is only performed on 5 pieces per lot. I understand the inspections to be CONTROLs of the process and not actual STEPs in the process.

I have asked for help from some of our customers but they seem to agree that inspections should be included in the process flow, but only if the inspection is performed on 100% of the product.

I would really appreciate some additional help in the decision of whether to keep inspections in the Process Flow Diagram or not.

Welcome ermoth! :bigwave:

In my view, each of the inspections should be included in their rightful places. They are control steps, but they are still part of the process.

I looked at this site (http://www.qualitytrainingportal.com/resources/fmea/fmea_10step_pfmea.htm)and concluded it's right to include inspection in PFMEA because it's possible to do inspections improperly and to assess the potential resulting consequences.

Welcome ermoth! :bigwave:

In my view, each of the inspections should be included in their rightful places. They are control steps, but they are still part of the process.

I looked at this site (http://www.qualitytrainingportal.com/resources/fmea/fmea_10step_pfmea.htm)and concluded it's right to include inspection in PFMEA because it's possible to do inspections improperly and to assess the potential resulting consequences.

:yes: Agree fully with Inspection included in all three...that's how I have always constructed my documents. Passed reviews with GM, DCX, Ford, and Toyota.:2cents:

ermoth - Welcome to the posting side of the Cove :bigwave:

We branch out the Inspection Steps as separate processing steps. The rationale used is that the inspection process, in and of itself, has Failure Modes which, we feel, need to be addressed (I guess that's pretty much what Jennifer said).

That being said, there is, of course, a "deviation from the norm" :bonk: . Our die cast machines have bad shot alarms. This is an in-line inspection of the process parameters and a "bad shot" is automatically (robotically) scrapped into a fenced off scrap tub. This type of inspection I do not break out but keep as part of the "Cast" processing step.

Hope that helps you out.

Thank you Jennifer, fuzzy and Bill!

I understand that Inspections have failure modes, such as "missed defect" or "missed inspection". But don't all "controls" have failure modes? What stops a person from including other controls that have failure modes?

"For example: Tooling Wear would be a potential cause of a failure, so we perform preventive maintenance on the tool. There are potential failure modes for the preventive maintenance itself, right? (inadequate cleaning, missed PM, etc.) Would you then include the PM itself as a step in the process flow? I guess I get to nitpicky with this type of thing."


I looked at the link (http://www.qualitytrainingportal.com/resources/fmea/fmea_10step_pfmea.htm) provided by Jennifer, and I see that each "component" of the process flow should have a function. "The function of the component is the value-adding role that component performs or provides." I specifically remember a registrar auditor telling me that inspections are not "value adding". So I guess that input leads me to believe that they should not be included as a step in the process flow.

Also, I took a crash course on the core tools last year and I remember the instructor saying that when you develop an FMEA, you should assume that each step before the current process operated correctly. In other words, only consider the failure modes that could happen at THAT process. What failure modes can occur at inspection if all the previous processes operated correctly?

What about the two different types of inspections I mentioned in my first post? IF an inspection is included as a step in the process flow, shouldn't it be an inspection that is performed on 100% of the product?

In my view, the sample size should not matter. I wouldn't want to make this more complex than it needs to be.

As for the value added question, I think it's open to different view points:

It's true that we should strive to reduce (and eliminate if we can) inspections through error-proofing the process, because the step does have costs that don't directly contribute to the product's sales. However, if the inspection is a control step it does have some value to the assurance of goodness.

Thus I maintain if the inspection step is important enough to perform, it should be represented in the planning and mapping of processes.

Welcome ermoth! :bigwave:

In my view, each of the inspections should be included in their rightful places. They are control steps, but they are still part of the process.

I looked at this site (http://www.qualitytrainingportal.com/resources/fmea/fmea_10step_pfmea.htm)and concluded it's right to include inspection in PFMEA because it's possible to do inspections improperly and to assess the potential resulting consequences.

:agree1: :agree1: :agree1: :yes: :thanx:
Andy

Thank you Jennifer, fuzzy and Bill!

I understand that Inspections have failure modes, such as "missed defect" or "missed inspection". But don't all "controls" have failure modes? What stops a person from including other controls that have failure modes?

"For example: Tooling Wear would be a potential cause of a failure, so we perform preventive maintenance on the tool. There are potential failure modes for the preventive maintenance itself, right? (inadequate cleaning, missed PM, etc.) Would you then include the PM itself as a step in the process flow? I guess I get to nitpicky with this type of thing."


I looked at the link (http://www.qualitytrainingportal.com/resources/fmea/fmea_10step_pfmea.htm) provided by Jennifer, and I see that each "component" of the process flow should have a function. "The function of the component is the value-adding role that component performs or provides." I specifically remember a registrar auditor telling me that inspections are not "value adding". So I guess that input leads me to believe that they should not be included as a step in the process flow.

Also, I took a crash course on the core tools last year and I remember the instructor saying that when you develop an FMEA, you should assume that each step before the current process operated correctly. In other words, only consider the failure modes that could happen at THAT process. What failure modes can occur at inspection if all the previous processes operated correctly?

What about the two different types of inspections I mentioned in my first post? IF an inspection is included as a step in the process flow, shouldn't it be an inspection that is performed on 100% of the product?

ermoth: I think you may confusing all the value stream mapping / kaizen / lean thinking information floating around out there with good old-fashioned APQP activities. I 100% agree with the advice you have been given by my fellow Covers:applause: and I find you are mixing up your objections to the advice across many topics. IMO the failure modes of your controls in PFMEA, should be considered as a whole when you determine the appropriate S- O- D numbers (Severity- Occurance -Detection). It is not an infinitely expanding chain of risks, as you examine every control process for it's failure modes, ad nausem:frust: . Nobody would ever finish a PFMEA if that were the case:( . My feeling is that perhaps you are over-thinking this issue...JMO:2cents:

The answers to the question of including inspection all have merit and point out that there is not a general answer that will suit all applications.
I do however have a concern. It goes beyond the question of flow charts & inspections, to the development and updating of FMEAs. I get the impression you are doing the upgrade on your own. FMEAs are risk analyses that require decisions to be made on the basis of knowledge coming from a team with varied input expertise. If my assumption is wrong, I hope it is, then your team will certainly have opinions (using the input from the responses) about how best to address the methods of testing in use.

RickT:
You haven't posted much, but when you do you hit a home run!:cool:

The point you make is so vitally important and, could be the very reason for the PFMEAs etc falling into the doldrums. It has happened so many times that the opportunity to get people together to create a PFMEA etc from collective point of view (and I don't mean a Borg collective, either), has been lost in the name of getting them done to satisfy a (customer) requirement.

Your cautionary comment is well timed/placed:agree1:

Andy

Thanks for the support Andy. I too have seen far to much of "single" FMEAs.
Live long and prosper!

RickT

Yes, I have a very bad habit of overthinking/overanalyzing. I just want it all to be very logical and clear, and I'm having difficulty doing that. I think my company will be sending several people to FMEA training, and I'm hoping that will help clear up the rest of my questions.

What are the failure modes for inspection? Missed inspection?

What are the failure modes for inspection? Missed inspection?

A couple we use all the time are "Good part rejected", "Bad part accepted". Depending on the inspection being performed there are some others our teams have come up.

A couple we use all the time are "Good part rejected", "Bad part accepted". Depending on the inspection being performed there are some others our teams have come up.

Ah, yes, the old Type I and Type II errors...simple roots might be seen as the correct criteria applied incorrectly, incorrect criteria applied correctly, etc., etc.

Good luck, ermoth...to strive for logic and purity in APQP is a good thing :D but as in all endeavours, there can be the law of diminishing returns...you can go crazy :( because what's right can be semi-subjective; great for one customer is seen as poor at another:frust: .

An alternative approach to the 'inspection' or 'test' function is to assess it's ability to properly detect (accept bad, reject good) as part of the DETECTION rating. (I use the Kappa test, Discrimination ratio approaches for repeatability and reproducibility; calibration for accuracy, etc.)

For unique tests, I would also look at work flow, etc. to assess and ensure appropriate controls to prevent mixing of samples, using a bad gage, etc. I do this to improve the overall inspection/test accuracy. These are things that wouldnn't often be caught in a measurement study. I guess I would cosnider this part of the FMEA (and one you can't do by yourself at your desk or as a team in a conference room - you have to observe it, or do some practice runs)

Then I would take the 'process' or handling of the inspection/test itself and assess how this could produce a harm. some examples: contamination from handling, a clamshell that could munch a part, etc. This is definitely part of the fmea in my mind

Yes, I have a very bad habit of overthinking/overanalyzing. I just want it all to be very logical and clear, and I'm having difficulty doing that. I think my company will be sending several people to FMEA training, and I'm hoping that will help clear up the rest of my questions.

What are the failure modes for inspection? Missed inspection?

1. Inadequate training of Inspector
2. Required check points not clearly stated in First Piece sheet as shown on Blue print / Customer specific.
3. Inadequate work instructions.

These are just a couple that I can give you, don't know if they help. I am in the same boat as you, learning more about PFEMA's everyday.

1. Inadequate training of Inspector
2. Required check points not clearly stated in First Piece sheet as shown on Blue print / Customer specific.
3. Inadequate work instructions.

These are just a couple that I can give you, don't know if they help. I am in the same boat as you, learning more about PFEMA's everyday.

FWIW - Just to throw my "slant" on these three - I would use them as Causes to the two Modes I suggested earlier.

FWIW - Just to throw my "slant" on these three - I would use them as Causes to the two Modes I suggested earlier.

me too.

failure modes are ways in which the function fails to function.
complete failure
partial failure
unitended failure
....

the function of inspection is to detect - and thus enable the removal of - defects.

so for me the failure modes are:

accepted defective material (escapes)
Damaged parts during inspection
rejected acceptable material

causes are wide ranging...