Answering a Finding that is Not? Work instructions are not always followed

I need some help with this one. I couldn't even come up with a good title. Here's the (redacted) finding:

Requirement: validation of specific methods & procedures (7.5.1)
Failure: Work instructions are not always followed
Evidence: ...load currents are below spec. Note this will not affect product , & subsequent Ac/dc for performance tests are validated after endurance.

Simply put, we demonstrated a test for the auditor and the reading was outside the specifications listed in the procedure. This occured shortly before our production folks left for the day so I really couldn't go back and get to the bottom of it: The next morning (after the audit is over and the auditor has left, of course) the unit in question was processed according to our non-conforming product procedures. There was nothing wrong. Further investigation (not by me, just the Team doing what they're supposed to do) found that a load cone (big resistive thing sort of like a light bulb) on the test bed had failed. The team replaced it. The unit (and the fixture) were put back through the test procedure, just like we say in our procedures. Everything was just like the specification.

How do I answer this finding? Everything worked like it is supposed to work. I think my production tech obviously gave some pretty bad answers during the audit as to what was going on or this wouldn't have generated a finding. I'll admit that I was brain dead after two days. I probably should have been quoting what comes next (fix problem, rerun tests).
 

Jim Wynne

Leader
Admin
Icy Mountain said:
I need some help with this one. I couldn't even come up with a good title. Here's the (redacted) finding:

Requirement: validation of specific methods & procedures (7.5.1)
Failure: Work instructions are not always followed
Evidence: ...load currents are below spec. Note this will not affect product , & subsequent Ac/dc for performance tests are validated after endurance.

Simply put, we demonstrated a test for the auditor and the reading was outside the specifications listed in the procedure. This occured shortly before our production folks left for the day so I really couldn't go back and get to the bottom of it: The next morning (after the audit is over and the auditor has left, of course) the unit in question was processed according to our non-conforming product procedures. There was nothing wrong. Further investigation (not by me, just the Team doing what they're supposed to do) found that a load cone (big resistive thing sort of like a light bulb) on the test bed had failed. The team replaced it. The unit (and the fixture) were put back through the test procedure, just like we say in our procedures. Everything was just like the specification.

How do I answer this finding? Everything worked like it is supposed to work. I think my production tech obviously gave some pretty bad answers during the audit as to what was going on or this wouldn't have generated a finding. I'll admit that I was brain dead after two days. I probably should have been quoting what comes next (fix problem, rerun tests).

What was the status of the unit tested in the auditor's presence? Was it assumed to be "good," or was it awaiting testing? If the unit was just a poor example (i.e., there was a problem that would have been identified during normal processing), I wouldn't answer the NC, but I would ask for the NC to be withdrawn after giving an explanation. If there was no actual nonconformance, just a misunderstanding, then there shouldn't be a finding. At the same time, and assuming my surmisal is correct, there might be an issue with the fact that the status of the unit wasn't apparent.
 
J

JRKH

Question...

The finding says that:

"Work instructions are not always followed"

What part of the instruction was not followed? If you ran a test for the gentleman and the item failed that in itself is not a failure to follow work instructions.

If the finding is based solely on this you could either contest it, or write some verbiage to the effect that you reviewed the process against the instructions and discussed the finding with the operator and found no problem.

James
 
Jim(s):
You are both headed in the same direction that I am. The unit in question was undergoing a 2 hour, full load, qualification test. We measure the load at the beginning of the test to make sure it is in specification and is really "full load". We measure again at the end to make sure the unit is still operating and providing power at the full load specification.
Jim Wynne said:
What was the status of the unit tested in the auditor's presence? Was it assumed to be "good," or was it awaiting testing? If the unit was just a poor example (i.e., there was a problem that would have been identified during normal processing).... At the same time, and assuming my surmisal is correct, there might be an issue with the fact that the status of the unit wasn't apparent.
The unit was "good" when we initially set up the test. We demonstrated the test for the auditor in the middle of the 2 hour run and found that the "load" was not in specification. So we basically determined the status of the unit as being questionable at that point. This problem would have been found when we checked the unit at the end of the test, per the procedure. Obviously, we didn't do a good job of explaining what comes next (i.e. treat product as non-conforming, determine cause, fix cause, retest product).

P.S. - I emailed a longer and more detailed explanation to the auditor before posting here. I am awaiting an answer. I posted this here for "consulting" since we wouldn't want to get that from a registrar's auditor (TIC). Also, I figured that this would make a good example to point to from the ISO/TS16949 Registration Maintenance Diary.
 
Last edited by a moderator:

Hershal

Metrologist-Auditor
Trusted Information Resource
A question here.....

If the requirement is validation of procedures, how does a fault during a test affect the validation? Would not the procedure have already been validated?

Hershal
 
G

Greg B

JRKH said:
The finding says that:

"Work instructions are not always followed"

What part of the instruction was not followed? If you ran a test for the gentleman and the item failed that in itself is not a failure to follow work instructions.

If the finding is based solely on this you could either contest it, or write some verbiage to the effect that you reviewed the process against the instructions and discussed the finding with the operator and found no problem.

James
Icy, I have to agree with Jim (and I think you support the same idea). The procedure was OK...you just identified a NC Product. How come this NC (from the registrar) was not discussed during a close out meeting and debated? Or was it? I hate it when registrars are ambiguous or just plain WRONG.
 

Ron Rompen

Trusted Information Resource
I think the real question here is; what Work Instruction wasn't followed?

Do you have a reaction plan for NC product? (Of course you do, it's a rhetorical question).

Does the operator know and (more importantly) understand the reaction plan? If so, did he follow it?

If he DID follow the RP, then there is no non-conformance....much as some registrars, auditors and (especially) consultants would like us to believe, having a QM system in place does NOT guarantee zero defects produced...it shoudl (hopefully) guarantee zero defects escaping to the customer.

If your operator did NOT follow the RP, then there is a non-conformance, in that the company failed to provide adequate training/tools/supervision to ensure that (s)he (gods, i'm starting to talk like a manager already) was able to properly perform all the tasks associated with the job.

I do agree with the others, however, that this should have been something discussed in the wrap-up meeting, and you should have been given a chance to respond then. Unfortunately, a lot of 'new' auditors are trying to win their spurs, due to the bad rep that QS had, and are being more stringent in the interpretation of the rules than they need to be (IMHO)
 

AndyN

Moved On
I'd choose the Forest Gump clause.......

"Stuff Happens" (or something like that). What I mean is, there will always be times when something fails. Not a fault of an ineffective calibration/maintenance program, or failure to follow procedures etc. It might be reasonable to call it a random failure, especially if there's no records of failures during test (do you record them as part of test protocols??)

Andy
 
Hershal said:
If the requirement is validation of procedures, how does a fault during a test affect the validation? Would not the procedure have already been validated?
It was either 7.5.2 c) Use of specific methods and procedures (i.e. If you set the tolerance too tight and then ignore it) or 7.5.1 e) the implementation of monitoring and measurement (i.e. it's not implemented if you don't use the tolerances. The procedure was not validated if you can't use the specifications to actually test the product.

Greg B said:
Icy, I have to agree with Jim (and I think you support the same idea). The procedure was OK...you just identified a NC Product. How come this NC (from the registrar) was not discussed during a close out meeting and debated? Or was it? I hate it when registrars are ambiguous or just plain WRONG.
This was debated at the close out meeting underneath the crushing weight of the following: the production team had gone home so we couldn't follow up, the procedural section in question had just been inserted the week before, I was not on top of my game at the end of my second 11 hour day, there was a witness auditor present, and I gave up since I was already going to have to change the procedure due to a lack of "special characteristics" symbols.

Ron Rompen said:
I think the real question here is; what Work Instruction wasn't followed?
None.

Ron Rompen said:
Do you have a reaction plan for NC product? (Of course you do, it's a rhetorical question).
Of course.

Ron Rompen said:
Does the operator know and (more importantly) understand the reaction plan? If so, did he follow it?
Absolutely.

Ron Rompen said:
If he DID follow the RP, then there is no non-conformance....
You are correct.

Ron Rompen said:
much as some registrars, auditors and (especially) consultants would like us to believe, having a QM system in place does NOT guarantee zero defects produced...it shoudl (hopefully) guarantee zero defects escaping to the customer.

If your operator did NOT follow the RP, then there is a non-conformance, in that the company failed to provide adequate training/tools/supervision to ensure that (s)he (gods, i'm starting to talk like a manager already) was able to properly perform all the tasks associated with the job.

I do agree with the others, however, that this should have been something discussed in the wrap-up meeting, and you should have been given a chance to respond then. Unfortunately, a lot of 'new' auditors are trying to win their spurs, due to the bad rep that QS had, and are being more stringent in the interpretation of the rules than they need to be (IMHO)
This was an unfortunate occurance of a failure in the test system rather than the product. Since this is rather rare, there was quite a bit of confusion about what was actually non-conforming, the product, the procedure, the obeyance of the procedure, or the test rig. Since it occurred about an hour before the closing meeting, there were time constraints on follow-up.

This is not a "new" auditor and I am not a "new" QM.
AndyN said:
"Stuff Happens" (or something like that).
That is why there is an appeals process.
 
In conclusion, Ladies and Gentlemen....

As noted previously, I emailed a description of this situation to my auditor. He followed it up with a phone call to me (now that's service). We discussed what happened and why and here is the result:
Icy's Email said:
Subject: NC3 has been rescinded
Dear Icy Mountain,
Based on our telecon, product was under test & in process. NC product procedure would have been next step, and my mis understanding.

The report will show only 3 NC rather than 4. Thanks for the clarification.
rjkoz

Thank you everyone for your advice and questions. The system works.
 
Top Bottom