When Must We Update a Control Plan?

[km2red]

Control Plans

We have individuals here who want to update the control plan every time something differs from what is going on, even temporary changes, or "tests" to see if the process can be improved.

The changes I am concerned about is when we are going above what the control plan requires (ie: adding an inspection). The inspection is NOT a perminant change (as of right now) it is more of an experiment to see if it will improve our process.

I say that the control plan does NOT have to be altered for this temporary change. The change can be communicated in other ways such as a quality alert, or other postings. If the change is determined to be needed or perminant, THEN the control plan needs to be changed (and other supporting documentation IE: FMEA's) to reflect this addition to the control of our process.

Any thoughts on the subject?

 

[noboxwine]

Re: Control Plans

km2red said:

I say that the control plan does NOT have to be altered for this temporary change. The change can be communicated in other ways such as a quality alert, or other postings. If the change is determined to be needed or perminant, THEN the control plan needs to be changed (and other supporting documentation IE: FMEA's) to reflect this addition to the control of our process.

KM2RED: YES :thedeal:

 

[Russ]

KM2RED:
I agree with you within reason. However, you must be sure that this doesn't get abused or you will end up with all kinds of document conflicts when people just do as they please and things quickly go awry! So use extreme caution here!

 

[Mike S.]

When doing an experiment, how about taking the existing control plan, having whomever is authorized make a handwritten change/addition to it (preferably in red) and also saying (writing) that this is a temporary engineering test to be done only on this batch, or on this day, or on this part (or some similar limitation)? That way, there is a record of what was done, and there are written instructions for the people doing the work, but there is no need for a formal revision of the control plan unless/until it is determined that the change should be made permanent.

 

[Claes Gefvenberg]

Time limit for the temporary changes

Hi km2red,

I agree that the control plan shouldn't have to be updated due to temporary changes.

Both Russ and Mike S make good points, however, and I would like to add one thing:

Temporary changes should be just that: Temporary. You may already be doing this, but I must stress the importance of setting a time limit for temporary changes. Our way of coping with this particular problem is to use temporary procedures, that can be valid for max 6 months (Usually less).

/Claes

 

[Bill Ryan - 2007]

When doing an experiment, how about taking the existing control plan, having whomever is authorized make a handwritten change/addition to it (preferably in red) and also saying (writing) that this is a temporary engineering test to be done only on this batch, or on this day, or on this part (or some similar limitation)? That way, there is a record of what was done, and there are written instructions for the people doing the work, but there is no need for a formal revision of the control plan unless/until it is determined that the change should be made permanent.

I don't agree. The CP does not have to be changed for every "Process Improvement" excersise you are undertaking
(isn't that the one of the uses of NCM tickets???)

The changes I am concerned about is when we are going above what the control plan requires (ie: adding an inspection).

In my experience, doing more then the CP calls for has not been an issue with our registrar (as long as it is "temporary"). Of course there are limits as to how far you can push the "temporary" limits.

I agree with Mike S. only if you issue CPs to the floor. Why not just use "temporary" Operator Instructions?

RUSS & CLAES also make valid points (at least at my company). If someone doesn't watch what is going on, suddenly, you've got a process change that doesn't get documented correctly (oops!!! - a finding on an audit - HEAVEN FORBID!!).

"Bottom Line" is - I don't think you touch the control plan unless what you are trying becomes permanent (the PFMEA is a different matter and I think the "Recommended Actions" column should probably be updated just to address the Continuous Improvement clause in QS/ISO).

Just my opinion (now I see why everyone uses "IMHO")

Bill

 

[Mike S.]

My point is this: When a process is running the "normal way", the "normal way" must be clearly defined by either training, documentation, or a combination of the two. If no documentation exists -- i.e. the "normal way" is defined only by training of the operator, somehow the operator has to be instructed what to do differently for the "experimental way". This could be done verbally but this is dangerous IMO since it is more likely to create a misunderstanding and there is less likelyhood that the actions and results will be clearly documented for future reference. This verbal stuff can start a "wild west" mentality of "anything goes". A simple note explaining what is to be done differently, with the time limitations, etc. signed by the responsible person is quick and easy and can help avoid many pitfalls and provides a record that should prevent any auditor from issuing a NC.

If there is documentation that spells out the "normal way", again something must be done to explain to the operator what the "experimental way" is. I don't condone a verbal change to a documented procedure even for a one-time experiment. Again, it is a simple thing to document what is to be done for the experiment either by marking-up the existing document that spells-out the "normal way" or writing a short, simple note regarding the deviation including the time limit and authorized signature. I think this is what Bill called "temporary operator instructions".

To summarize: Describe the change in writing (somehow) and, as we all seem to agree, clearly establish the time limit. JMHO.