Increase in Major TS16949 Audit Findings

[MWR4QA]

Fellow Quality professionals,
Has anyone noticed an increase in major audit findings in the past six months? I am concerned about a particular way in which the nonconformances are being found.
During a surveillance audit, the auditor issues a minor NCR for the calibration system. Let's say she found two MTE's in production whose calibration dates were exceeded. During the next surveillance audit, she finds two more MTE's that are out of cal. The IATF rules require her to issue a major nonconformance against the calibration system and the corrective action system.
My concern is that these minors are being turned into majors in areas where there are many (1000's) of opportunities for error, such as calibration, material identification and adherence to Control Plan. There is no major system breakdown, we are just not able to assure 100% conformance.
Is anyone else seeing this?

 

[Sidney Vianna]

For what is worth, in the AS91X0 world, the IAQG has a similar requirement. AS9101 requires the following

Recurrence of the same or similar nonconformity found during consecutive audits at a particular site/location shall be considered as a failure of the corrective action process (see 9100-series standards clause 8.5.2) and shall result in a major nonconformity being issued.

When I was involved with the ASQSC, I voted against the revision of that document (AS9101) because of that.

The process HAS to allow, as you mentioned, the analysis of the shear number of opportunities for error/nonconformities. There is no perfect system in the world and to mandate automatic major NC's (with all it's implications) in case of recurring similar nonconformities is unwise, in my opinion.

 

[Ninja]

...During a surveillance audit, the auditor issues a minor NCR for the calibration system. Let's say she found two MTE's in production whose calibration dates were exceeded. During the next surveillance audit, she finds two more MTE's that are out of cal. The IATF rules require her to issue a major nonconformance against the calibration system and the corrective action system.

Minor or major aside (though I know that there is a HUGE difference between them)....let's look at the example...

The purpose of having a calibration tracking system is to ensure that all gages being used are in calibration period.
The purpose of a Corrective action system is that when an issue is recognized, it gets corrected.

Gages were not in claibration...the system ain't doin' what it was designed for.
A requirement is then placed to correct it.
...and it wasn't corrected...the correction system aint doin' what it was designed for...and the calibration system still ain't workin' either.

What should the response regarding a system that ain't workin'?
After it was found to not be workin' and after a period still ain't workin'?

If a system is built to accomplish something, and it doesn't accomplish it...that's a pretty major thing, no?
And if the safety net and pieces picker-upper isn't working either...it's a sort of huge thing, no?

I would be interested to understand the rationale behind and downgrade of finding level.
I understand from the above "It's hard, so it shouldn't have to be working all the time"....and I'm quite sure that isn't what you intended to communicate. Please help me understand what you mean...

 

[Sidney Vianna]

Please help me understand what you mean...

have you ever been responsible for a system with thousands of documents and managed to make sure no obsolete document was ever being used?

have you ever been responsible for a system with thousands of gages and managed to have all of them recalled and re calibrated before they were overdue?

A quality system is no guarantee of perfection. It is foolish to think a large and complex system with many opportunities for failure will not have occasional nonconformities.

 

[JoShmo]

Being past a recall date and being out of cal are 2 totally different things and one has greater potential for problems than the other. Why don't you have a system in place to allow for use of equipment past the published/indicated due dates. You may not be helping yourself to avoid such issues - looking for a problem with the number of audit findings and CB rules isn't going to assist you at all. Fix the issue, not a sypmptom.

 

[Ninja]

have you ever been responsible for a system with thousands of documents and managed to make sure no obsolete document was ever being used?

Yup. Live there now. Docs are published electronically only if status is "Distributed"...Doc Coordinator changes status to "Obsolete" (or anything other than "Distributed") and it disappears from everyone's access privileges.

Was on paper...obsolete docs in use were common...then we corrected it.

have you ever been responsible for a system with thousands of gages and managed to have all of them recalled and re calibrated before they were overdue?

Yup. Live there now. It has issues and we are creating a better way.

My questions:
- Have you ever had a customer or an auditor find a system that was not working right and decide that it was OK?
- Have you ever had a customer insist that you correct something, and accept later that you didn't correct it?

I have never had a car that didn't work and consider it OK. I repaired or replaced it.
It works most of the time doesn't get me to work.

I've never considered it OK when flipping a switch didn't turn on the lightbulb. I changed the bulb or rewired the circuit so that it performed as intended.
It works most of the time doesn't let me see.

If it is OK to have overdue gages on the floor in use...then write that into the governing doc that it is OK...then you have no finding.
If it isn't OK enough to rewrite the doc...because it is "not OK" to have such gages in use...what is the issue with an audit finding?

It isn't that these things are easy, or that things are perfect all the time...
but why take issue with a repeated, justified finding?

 

[Sidney Vianna]

I can guarantee you. If you have a system with thousands of documents/command media pieces, someone is using a print-out (hard copy) of a document and, chances are, a few will be obsolete.

but why take issue with a repeated, justified finding?

You don't understand, apparently. It is not a problem to take an issue with a justified nonconformity. You identify it, you correct it and you attempt to prevent it from happening again, via the corrective action process.

But when you have a system with a huge number of opportunities for failure and you mandate an automatic major NC in case of similar nonconformities, the auditor will have to come back and scrutinize the system again, in depth. If s/he finds another (single?) nonconformity, then s/he must initiate de-certification proceedings. Is that what we want?

If you have 10,000 gages in your system and 5 are being used, but overdue for calibration, does it mean that the system is broke? Keep in mind that, just because they are overdue for calibration, it does not mean that they are out of tolerance, obviously.

Corrective actions are supposed to fix systemic failures. The one offs will happen even when we have a system. The price of perfection would probably put most organizations out of business.

 

[Ninja]

If you have 10,000 gages in your system and 5 are being used, but overdue for calibration, does it mean that the system is broke? Keep in mind that, just because they are overdue for calibration, it does not mean that they are out of tolerance, obviously.

LOL, ... I think we are "Agreeing violently" again...it wouldn't be the first time for you and I.

In the above quoted situation, there are multiple ways to "correct" it.

- One is to be perfect, and I agree that the cost of that can easily be prohibitive.
- One is to make allowance for these situations in the system itself...essentially building the system to function in a less than perfect world.

If "Past calibration recall date" is going to happen...handle it in your system.
If "Past calibration date is not necessarily a deal killer (which I think we agree it is not)...then state that in your system.
If "Out of tolerance" is the deal killer (which I think we agree it IS)...prescribe what the proper action is when a gage is found "Out of date" but not "Out of tolerance".
While you are at it, prescribe/define the proper response when a gage is "Out of date" AND "Out of tolerance".
By doing this, there is no NC for a date on a sticker when the gage is still OK...and thus no finding...and thus no escalation.
{For that matter, the sticker itself is not required in the first place}.

Things can escalate when you've painted yourself in a corner...get out of the corner.
The issue isn't the escalation on repeated NC...it is the NC that could have been avoided easily and wasn't.

 

[mike_o]

LOL, ... I think we are "Agreeing violently" again...it wouldn't be the first time for you and I.

In the above quoted situation, there are multiple ways to "correct" it.

- One is to be perfect, and I agree that the cost of that can easily be prohibitive.
- One is to make allowance for these situations in the system itself...essentially building the system to function in a less than perfect world.

If "Past calibration recall date" is going to happen...handle it in your system.
If "Past calibration date is not necessarily a deal killer (which I think we agree it is not)...then state that in your system.
If "Out of tolerance" is the deal killer (which I think we agree it IS)...prescribe what the proper action is when a gage is found "Out of date" but not "Out of tolerance".
While you are at it, prescribe/define the proper response when a gage is "Out of date" AND "Out of tolerance".
By doing this, there is no NC for a date on a sticker when the gage is still OK...and thus no finding...and thus no escalation.
{For that matter, the sticker itself is not required in the first place}.

Things can escalate when you've painted yourself in a corner...get out of the corner.
The issue isn't the escalation on repeated NC...it is the NC that could have been avoided easily and wasn't.

Let's try a different example (since calibration audit findings are pretty much the domain of less skillful auditors).

In a manufacturing organization of several hundred associates (jobs are low to mid-level in difficulty and the education/skill level is often less than high school graduate) - a minor n/c is written for a container of in-process material that is missing a label. It could be that the operator neglected to apply the label, or it may have fallen off. Each item of product in the container is identified and status is clear from location of product within the process. We agree this is a minor n/c and the corrective action was a revision to the work instruction for how the associate is to react when an unlabeled container is received at their work station.

Next audit - different associate in a different building, different product line. Associate is required to inspect every 20th piece. Auditor counts the number of finished pieces, divides by 20 and determines that the total number of checks performed does not match the requirement (although evidence indicates checks were done throughout the production run). This is written as a minor n/c BUT the auditor has determined that it is a repeat of the previous minor n/c because he has written it against the same clause in the standard. Therefore he ALSO writes a MAJOR n/c for failure of the corrective action system, even though the two issues are totally unrelated.

In this environment and with this level of workforce, you will invariably find someone who has decided to take a shortcut. Human nature is not a system failure, it is an inescapable reality. With the approach the auditor is taking, we are looking at the potential for a perpetual string of minor/major nonconformances, thereby guaranteeing the registrar extra on-site visits and the promise of the requisite fees that go along with them.

Opinions are welcomed.

 

[JoShmo]

I, for one, are happy to see your comments about calibration and auditors - something I was making a point about but got a different answer...

TBH mostly stuff like this is just a symptom. It's easy pickins. Find a couple and make it a big deal and it don't matter if it is or not. There's no attempt to link it to cause/affect. As you say, the parts were clearly where they are and what they are. So did the auditor on finding the second situation, then change tracks and ask about if the internal audits saw anything? Has there been a change to the process/people/label supplier or who know what else? probably not. Too easy to make the company do the leg work because they don't know any better, because they only do audits once or twice a year so arent any smarter and the management review is a dog and pony show while the rest of the plant staff sit and doze... But that's just my opinion and a dose of reality...