Engineering retest in line - Machine stability issues

[Graciel]

The company I work for has a lot of problems in machine stability (sometimes if you pass again and again same part in the machine, it creates different results, sometimes approved, sometimes rejected). This is causing some issues on production line non conformity rate, since it can measure 4% of reject in the first place, but latter if you retest the parts, the result changes to 3%. This data sometimes is a result of engineering parameters change. Here the software engineer increases the parameter range frequently because they say they are re-conditioning the machines. This type of change, was said not to be considered part of process change management, since is known by the high management that this will happen frequently. If in some point of a IATF 16949 audit, the auditor gets to this matter, what can he say? Can this be an issue?

 

[Ninja]

Hello Graciel,

That sounds like the situation that MSA Gage R&R was created to address. Have you performed this study on your test method?

 

[Graciel]

It was performed by quality department at the beginning but they say it was OK. Now they are doing it again for the part with the worst rejection rate only. But my concern is during an audit what should be explained and what is an auditor wanting to see if he/she gets to this matter. Since is impacting on NC rate KPI (there are parts being rejected that later is approved, after machine adjustments).

 

[Bev D]

If you are changing the accept/reject limits (as you seem to be saying) and you are not handling this change as a change to product quality then an auditor will have major things to say about that.

 

[Ninja]

Here the software engineer increases the parameter range frequently because they say they are re-conditioning the machines.

IMO (not an auditor), the need to change the parameter range due to reconditioning should take the machine out of use until it is re-tested (preferably by GRR) to be like was originally approved or better.

As an auditee, I would put my armor on and expect the auditor to take a very deep dive into this and write a finding.
...or you can call out before the audit and address it as a "self caught" NC ... that would go much easier in the long run.
Software engineers don't get to write specs at whim...(unless, of course, they are software engineers from your customer...)

 

[Graciel]

IMO (not an auditor), the need to change the parameter range due to reconditioning should take the machine out of use until it is re-tested (preferably by GRR) to be like was originally approved or better.

As an auditee, I would put my armor on and expect the auditor to take a very deep dive into this and write a finding.
...or you can call out before the audit and address it as a "self caught" NC ... that would go much easier in the long run.
Software engineers don't get to write specs at whim...(unless, of course, they are software engineers from your customer...)

I thought about registering it as internal NC but the problem is, what they are doing (they, I mean Quality supervisor, machine engineers, production manager, high management): they are signing internal deviations for it, but they register it as "not necessary to warn customer". If I show this to an auditor, it can make things worse, right?? So if I put this in my Internal NC action plan, I´d be complicit with this whole process.

 

[Ninja]

Well...a question or four first...

Who are you (position)?
What is your relative position to the Quality supervisor?
Have you discussed this with him/her? (answer should be yes here)
What was their take on it?

We are on a forum...knowing only the info you type...I choose to believe that the QS knows a bit more about your line, and your relationship to the customer.

 

[Graciel]

I am system management analyst. Yes, I have spoke with him, he suggested the internal deviations. Also, you can ask: so who´s your boss? (it´s the administrative manager, he doesn´t have experience with system management).

 

[Ninja]

Understood...thank you.

If it were me, I would document the conversation with the QS for my own records and then see what the audit brings.
I might (socially, not professionally) ask the QS if he/she thought his/her approach would stand up through an audit, but this strikes me as one of those cases where you have to let people learn from mistakes if they don't react the way you think they should when you ask a pertinent question.

...depends on your relationship with the QS...