Loss Prevention in ISO 9001:2000

[Denis]

ISO 9000:2000 loss prevention

Hi,
I'm new to this group , so apologies if this has been previously explored.
We are working towards ISO 9000:2000 standard - already approved for '94.
Need opinions on how we are to address the loss prevention issue, section 8.5.3. Our business is semi conductors manufacture and we feel that conventional FMEA just will not work due to the infinite number of variables etc..

Any ideas on how we can go about addressing this part of the standard?

 

[km2red]

How about a "generic FMEA"? One that catorgorizes one TYPE of product, but not every specific part number? (like a product family)?

 

[Jim Biz]

Contensious thoughts!!

Welcome to the group Denis

Guess I've often wondered why a number of folks really believe that FMEA (in one form or another) is mandated by the standards.

From your post - I would think many of the other "suggested" quantatiative data areas in 9004 such as market analysis/ process measurements/ self analysis data ( in house Cost of Quality data trends ) etc., could be used - in lieu of "formalized FMEA"

One local area presentation that I attended - led attendees to believe that there was frankly "NO OTHER way to satisfy the issue (this was in connection to the presenters offer to teach FMEA classes for those present.)

My thinking is that because the wording on 9004 - uses language terms of "SUCH AS" fault mode and effect analysis on the first suggestion line that everyone quits reading at that point.

 

[Marc]

Motorola uses FMEAs in the semi-conductor sector quite effectively. I'd post one but obviously that wouldn't be appreciated by them. I'm not sure why you:

> feel that conventional FMEA just will not work due to the
> infinite number of variables

In addition, FMEAs are only one avenue of addressing 8.5.3 - Or are you looking for a better tool because you're focusing on preventive action in your manufcturing only.

Remember, in whole the requirement reads much like this:
*****************
8.5.3 Preventive Action

Our organization determines action to eliminate the causes of potential nonconformities in order to prevent their occurrence. Preventive actions are appropriate to the effects of the potential problems.

A documented procedure is established to define requirements for

a) determining potential nonconformities and their causes,
b) evaluating the need for action to prevent occurrence of nonconformities,
c) determining and implementing action needed,
d) records of results of action taken (see 4.2.4), and
e) reviewing preventive action taken.
*****************

As a last comment, if you're just trying to fulfill 8.5.3 - do a search here in the forums using the key word 'preventive'. There are a bunch of threads where we discuss preventive action.

 

[Marc]

Re: Contensious thoughts!!

Originally posted by Jim Biz

Guess I've often wondered why a number of folks really believe that FMEA (in one form or another) is mandated by the standards.

This is because, if you look closely, you will see that the ISO 9001 revision was strongly influenced by QS-9000. If you take a trip back into time to The Old Forum Archives (Note: Clicking the link will call up a pretty big - 350k - page) - and even within these newer forums - you will see that addressing the requirement in QS-9000 for predictive methods was, as a minimum, fulfilled by FMEAs.

Another 'preventive' issue is: If you react to a corrective action and part of that reaction is to review other products / processes where that 'error' may occur in, is this really a preventive action. I think the current agreement is that this is preventive even though the origin was from an 'error'.

Between auditors and registrars who are involved in both ISO and QS the 'recommendation' for an FMEA comes out. This is in part because of the limited number of tools which are predictive. Just to recall how pretentious the issue gets when you come to distinguishing predictive vs. preventive, remember this thread? http://Elsmar.com/Forums/showthread.php?t=687

This also came this weekend:

> Hi Marc,
>
> Thanks for your reply on loss prevention issue. The major difference
> between Motorola and us is that we actually grow the compound semi-
> conductor wafers for customer end use and conversion into workable devices.
>
> We effectively have a one step manufacturing process (growth of epi
> taxial wafers) completed in one step, controlled by Mass Flow Controllers
> via computer program.
>
> We have talked about performing FMEA on the actual processes
> involved in all the manufacturing stages as opposed to FMEA on the product.
>
> Our customers supply us with their design specifications and we make the
> wafers to their spec.
>
> Ultimately we are looking for a better tool to achieve a loss
> prevention mechanism.

The bottom line is you're looking for a predictive risk analysis tool to address loss prevention - a predictive methodology. In considering the FMEA I prefer a procsss FMEA as opposed to a product FMEA. In you case, I think a product FMEA would be a pain. I would insist on process FMEAs.

OK gang - Ideas on Process Loss Prevention Tools? I did some searches here and there's virtually nothing specific - other than with respect to FMEAs...

 

[Marc]

Originally posted by km2red

How about a "generic FMEA"? One that catorgorizes one TYPE of product, but not every specific part number? (like a product family)?

How about a process / operation FMEA? Is this what you have in mind?

 

[Denis]

Thanks for the comments so far.

The most useful idea that we are now talking about is using FMEA on the process, coupled with an internal preventative and corrective action course.

I know that we would still be reacting after the event on some of the issues, but at least we can demonstrate our commitment to providing permanent corrective actions.

Hopefully, customers and third party audit bodies will recognise this.

Regards,
Denis