Paragraph 4.14.3 b) - When is a Preventive Action required?

[CarolX]

Can someone help me with this section of the standard. I am confused as to many things.
Example:

Paragraph 4.14.3 b) states "determination of the steps needed to deal with any problems requiring preventive action"

My definition of "preventive" was to take action before a problem occurred. I.E.: Maintainence on equipment, calibration of gages, inspection of raw material before use are some of my examples of preventive actions. Taking action to "deal with any problems" is corrective action in my book.

What am I missing here - any help????

Thanks,

CarolX

 

[Jim Biz]

Here's a place to start. Don't be worried about confusion with this one!!

http://Elsmar.com/ubb/Forum31/HTML/000023.html

If you are familliar with D-FMEA or P-FMEA conducted Prior to a problem occuring and/or 8D 5Y after a reported defect occurance... it will help give you a better idea of the prevent - correct dilema..

Regards
Jim

[This message has been edited by Jim Biz (edited 20 July 2000).]

 

[CarolX]

WOUCH - I can see I am not the only one to ever have problems with this one. Thanks for the lead - truly appreciated.

I will continue to hash through this one - should be fun (NOT)!

Regards,
CarolX

 

[Bryan]

Preventive Action or Not

The may be straight forward, but we have had some discussion here so I wanted a outside opinion for clarification.

Example: Parts are rejected from the mfg floor for a nonconformance, some problem with them. Its not a serious trend right now, the rejection of this part. The question would be, Can a Preventive Action be written to have this part that remains in stock, checked and if necessary rejected, so no more are issued to the floor to prevent further rejects which slows production??
Or is this a case that a CAR should be written ??

Thank you for your input
Bryan

 

[Jim Biz]

Just IMHO:
I'd call it corrective: it already happened wether or not it got shipped out or not it was not prevented before it occured.... if/when there are other similar products that may show the same condition in the future then I'd call those actions preventative.

Guess I'm operating on the assumption that the error in reality-found-you or occured prior to a preventative application in place ... that's corrective - any action taken prior to an actual error occuring is preventative

Regards
Jim

 

[Al Dyer]

Jim,

Agree that this instance is a corrective action.

Shouldn't the initial corrective action also identify like proceeses/products that could be affected and therefore corrected?

If the same problem occurs again in the like process that has been corrected, wouldn't that then become another corrective action and open up the effectiveness of the original corrective action.

I think we are on the same wave length:

Corrective Action - Occurred
Preventive Action - Could Occur (FMEA?)
Continuous Improvement - All's well, let's improve it

ASD...

 

[D.Scott]

I sure don't want to get hung up on semantics here and basiclly I agree with Al but I don't think his definitions go far enough to seperate Prevention and Continuous Improvement. To be a Prevention, it must have the same "all's well" characteristic as the C/I. To me the big difference is that the true Prevention has no measureable or goal (other than 0 which has been the state all along - otherwise you would be doing a C/A). There is never an occurance if the Prevention is successful. On the other hand, every C/I has a goal and a measureable (or else it would be called a "Continuous Status-Quo").
Just my opinion but it seems to make life easier here in tracking and the auditors have never raised an objection.
Dave

 

[Al Dyer]

Originally posted by D.Scott:
I sure don't want to get hung up on semantics here and basiclly I agree with Al but I don't think his definitions go far enough to seperate Prevention and Continuous Improvement. To be a Prevention, it must have the same "all's well" characteristic as the C/I. To me the big difference is that the true Prevention has no measureable or goal (other than 0 which has been the state all along - otherwise you would be doing a C/A). There is never an occurance if the Prevention is successful. On the other hand, every C/I has a goal and a measureable (or else it would be called a "Continuous Status-Quo").
Just my opinion but it seems to make life easier here in tracking and the auditors have never raised an objection.
Dave

Dave,

I agree, I did not go far enough into the thought process behind my comment. Here goes,

A hypothetical situation #1:

I use FMEA review and prioritize the RPN's for preventive action projects.

Do I have a measurable? Yes, the reduced RPN (if successful)

Do I gave a goal? Yes, the lowest RPN possible.

Can I have an RPN of "0"? Common sense says no, no process is perfect.

Hypothetical #2:

I am running a daily chart on out internal PPM. The goal (limit) is 25.

1: My daily reading for the last month show I am consistently running in a range of 5-10 PPM. I say, people we can do better. Let's start a "continuous improvement" project.

2: Same as month 1 (5-10 PPM). At the start of month 2 we start to see the daily PPM's creeping up towards 13-18. I say, people, let's start a "preventive action" project. My reasoning is that we are within our goals but we want to prevent the PPM from going any higher.

3: Same as month 1 (5-10 PPM. I look at yesterdays data and see that the PPM is at 50. I say, people, shut down the process and begin a "corrective action" immediately!

I think any auditor worth their salt will see that there is a fine line between preventive action and continuous improvement. If a company documents that they are acting upon their defined measurables and improving their processes and systems there should be little discussion between the fine line between the 2 activities.

ASD...

 

[Marc]

Originally posted by Al Dyer:

Jim,

Agree that this instance is a corrective action.

Shouldn't the initial corrective action also identify like proceeses/products that could be affected and therefore corrected?

If the same problem occurs again in the like process that has been corrected, wouldn't that then become another corrective action and open up the effectiveness of the original corrective action.

I think we are on the same wave length:

Corrective Action - Occurred
Preventive Action - Could Occur (FMEA?)
Continuous Improvement - All's well, let's improve it...

Also see: http://Elsmar.com/Forums/showthread.php?t=3179

If your question is:
->The question would be, Can a Preventive Action be written
->to have this part that remains in stock, checked and if
->necessary rejected, so no more are issued to the floor to
->prevent further rejects which slows production??

In so far as whether or not to request a corrective action in this case you have to look at things holistically. If you got bad parts in and it is a rare case - never happened before or maybe it happened last year but a different supplier (or maybe even the same supplier) - and it's not serious (such as it doesn't shut down your line or something) then you probably will not want to ask for a CA. Use common sense. Your system should allow you to write up a nonconformance report which will trigger containment (which would include a stock purge) without requiring a corrective action.

Corrective actions take time and cost money. You ought to have a good reason for one before you request one.

The logic ruling is:
Every corrective action is the result of a nonconformance but not every nonconformance requires a corrective action unless your system was designed so tight it's beyond what makes sense (and it squeaks). I have seen companies drown in corretive actions when the system was designed so tightly that every nonconformance required a formal corrective action. Dumb.

 

[Marc]

Also see: http://Elsmar.com/Forums/showthread.php?t=3186
and
http://Elsmar.com/Forums/showthread.php?t=3171