Hello All,

I am looking for some opinions from all you fine folks here. What do you use as a guideline and/or requirement for establishing a formal corrective action request? I mean, like the grey areas. For example, a customer rejects 2 pieces out of a 1000 part order, and the parts are a penny apiece. Not an effective use of resources to issue a CAR for this.

Any thoughts?

Carol X

Carol,

I think that you could probably very easily justify not performing corrective action according to 8.5.2.c evaluating the need for action....

What I would do is investigate, (albeit breifly) and document why we decided that the risk involved was insignificant compared to trying to correct the problem. But, if you have 2 bad parts in every 1000 you send, you might rethink your position.

I hope that is what you were looking for?

Carol,

Are we talking ISO, ISO/TS, or QS?

At the least it has to be to the "degree appropriate" and commensurate with "risk". For me the first step would be to review the FMEA to determine degree and risk.

I would probably lean towards a corrective action that might say the root cause is undeterminable and that the C/A will be closed in 6 Mo. (?) if the situation does not re-occur. If it does happen again then there is a definate problem that needs to be resolved.

I guess I'm paranoid of PPM's hovering around 2000.:ca:

Carol,

I copied portions of this thread --> http://Elsmar.com/Forums/showthread.php?t=3188 to address exactly what you are posting about.

Thanks for the responses...but what I am trying to develope, and hope to get some input from all by example, what is the threshold you use to initiate a corrective action. Is it based on the percentage of a "lot" rejected, or based on a dollar amount? My current procuedure was written very vague (intentionally on my part to give me latitued(sp)), but now the big cheese wants some more definitive guidelines.

From SteelmaidenI think that you could probably very easily justify not performing corrective action according to 8.5.2.c evaluating the need for action.... see above...thanks for the input, already doing it this.

From AlAre we talking ISO, ISO/TS, or QS? ISO9002-1994I guess I'm paranoid of PPM's hovering around 2000.

PPM-Not an issue here

Energy,

Thanks my friend...not quite what I was looking for. I guess I am just trying to do some benchmarking. Just gonna do some more of this:frust:

LOL. Thanks!!!

CarolX

If the cost of those original 2 pcs exceeds your production rate of $100 ($50 per hr for your salary and $50 per hour for all the company benefits, write it up. If not, file it somewhere and monitor for repetition.:rolleyes: :ko: :smokin:

4.7 Corrective Action
4.7.1 If permanent corrective action is required to prevent future occurrences of the issue which prompted the MDR, the QA Auditor shall include the RCA # assigned from entry into the Corporate RCA Database.
4.7.1.1 If a Plant process is involved, refer to CWI-14-QAD-0001.

Hope it helps, Eileen

Originally posted by energy
If the cost of those original 2 pcs exceeds your production rate of $100 ($50 per hr for your salary and $50 per hour for all the company benefits, write it up. If not, file it somewhere and monitor for repetition.:rolleyes: :ko: :smokin: Energy - you are killin' me......$50/hour...I am DEFINITLY gettinig way under paid!!!!!:vfunny:

Eileen,

4.7.1 If permanent corrective action is required to prevent future occurrences of the issue which prompted the MDR, the QA Auditor shall include the RCA # assigned from entry into the Corporate RCA Database.

Not unlike my procedure....enough to satisfy the requirements, but vague enough to let you manage it properly.

And that's my quandry...the boss wants hard, fast guidelines, i.e. a detailed set of instruction on when exactly to initiate a documented corrective action.....which I fear will back us into a corners, but..... :bonk: Life goes on!

Carol

E-Wall,

If a permanent corrective action is needed I would think that a corrective action has been initiated. No?

Hard to tell from the submitted snipet, I don't know what your MDR is, I guess it depends on your entire procedure/work instructions.

Just fishing for info!


Carol,

Example

A formal corrective action is issued when:

- A request for C.A. is submitted by the customer (Int. & Ext.)
- Ongoing scrap rate is above targeted scrap rate
- Process is below targeted capability goals (Cpk/Ppk)

For a small problem many companies opt to use a "Heads Up" issued to the production floor and temporarily increase spot inspections. The heads up keeps the operator in the loop and hopefully increases the chance of finding additional defects, if they exhist.

Al (and Carol too really): Temporary Corrective Action - We are human and even though you may have a good procedure in place...folks tend to get away from all the details if/when they rely on their memories rather than the provided reference documentation. If this is the reason bad product got out the door (rather than insufficient process) to us that calls for temporary corrective action measures. To us this is short term additional inspection requirements, monitoring of product or process by a third party to ensure process (currently in place) is followed all the time, etc...

Permanant Corrective Action - When we actually change the process, equipment, requirements, documentation, etc...

MDR for us is a Material Discrepancy Report. This is used internally only and how we track all material nonconformity issues regarless of Corrective Action (temp or perm) or just BS Customer Returns (product tests fine - but close to the 1 yr mark sitting in their warehouse...you know the political/scratch-my-back sales game).

A RCA is only generated for permanant Corrective Actions and is ultimately the call of the QA Manager although the information is reviewed by Engineers, Material Mgmt, and Mfg folks along with QA and the input reviewed by the QA Manger. Issues taken up with Plant Mgr if there are dissagreements on how it should be handled (very rarely needed).

Carol: Hard and fast guidelines for us is the 'rule-of-thumb' as I have stated in this post. Basically, if the root cause is training (aka - human error) issue rather than a failure of the process or equipment, we don't write it up as a corrective action.

Also, here is another snippet:
1.1 The purpose of this document is to ensure that timely and effective corrective or preventive action is taken to prevent, reduce, or eliminate non-conformances which degrade the quality and value of the product.

1.2 Request for Corrective/Preventive Action may be initiated as a result of nonconformities relating to process and product, audit findings, improvement teams, management review meetings, or review of standard quality records.

E-Wall,

I believe we're talking apples and oranges. Training and human error are not synonomous but can be related depending upon the dynamics of the organization.

Human error is going to happen annd can be reduced by giving people the tools and processes they need to perform and take accountability for their performance. Error or mistake proofing come to mind as actions when human error is detected.

A person can be well trained and still make an error, as in Q.A. performing a 100% inspection. No way to catch everything.

Human Error as a root cause: Good luck if you supply automotive and have an SQA with some time under their belt.

Training as a root cause: Why would you not issue a C/A if your training program is not effective? The company should take credit for identification and correction of the problem.

Whatever the situation, effectiveness is the ultimate goal.:bigwave:

I understood the question to be about generating Customer based Corrective Actions (that are actually seen by the customer)and that is how my response should be viewed. We have a seperate Plant level system (internal) where we do issue Corrective actions - regardless if it is human error, training, or whatever. There was also a seperate Supplier system, but we won't go there now.

Also, you can only blame training or human error for so much...and this way if we have a recurrance of a situation then there has to be a change to the process - because obviously it cannot be maintained (regardless if it is a people issue or not) by the folks doing the work so the process must change to be effective in spite of the people working it.

An no, we don't supply automotive. I worked at a machining plant prior to working here and what a shock when I was used to the QS system to see how ISO 9002 worked! I still tried to sell the benefit of some components of the QS system - but there was no way they were buying in (the first year here I called it the 'dark ages').

As I have said in the past, I've worked here for 4 year now, and change is slow to happen...but I have been seeing the payoff (of various levels, naturally) for the last 2 years. We still have more system maturing needed, but that is the ultimate continual improvement...and as long as we don't backslide, I will be happy (but not content or less vigilant).

What we have in place works for us (also doesn't mean that we are not trying to improve), that doesn't mean eveyone will like it or want to use it...so I don't expect anyone here to throw me a parade for sharing these jewels of information... :vfunny: And I keep learning too by others offering what is done in their systems. After all, isn't that why we all come here?

Eileen

PS: Just before Christmas the Corrective action Systems (Customer, Supplier, and Internal) procedures were 'combinded and standardized' into one process but I haven't seen it in practice yet, so can offer no information. This process has been in flux since the person who has handled it for over 3 years was laid off. We also were pretty much shut-down for the last week of Dec and in some areas the first week of this year.

There have been numerous changes to both personnel and systems that have not been communicated effectively. This is one of my gripes.

Originally posted by CarolX

I am looking for some opinions from all you fine folks here. What do you use as a guideline and/or requirement for establishing a formal corrective action request? I mean, like the grey areas.The problem I have seen is where a company ends up writing a Request for Corrective Action for every nonconformance that pops up. Then the company ends up with a mess and unhappy employees.

Typically I ensure clients have a 'Gate' defined in their system. Every nonconformance is entered into a list or database (company size factor), but no request for corrective action is issued until the 'Gate' approves.

Now - how does the 'Gate' approve or disapprove? The 'Gate' person looks over the request, does a search (or whatever) to see if it is a repeat, considers the over all effect (etc., etc.) and makes a (now you're gonna hit me) 'Common Sense' decision based upon all the information at hand. Typically the 'Gate' person is the QA Manager or someone who understands the operation from a broad view. This includes RFCA's from customers - not every Request for Corrective Action from a customer is valid as we all know. I have also seen this done in automotive situations where the Gate determined that the customer request for a formal CA was not valid. The Gate notifies the customer it is invalid and they fight it out from there, so to speak.

If the 'Gate' determines there is a NEED for a formal CA, it is entered into the Corrective Action system and the process is begun. If the Gate determines a formal Corrective Action is not needed then s/he gives their reasoning and closes out the record.

This, in some companies, is the same way a determination is made as to whether a nonconformance is really a nonconformance or not. The Gate decides, for example, whether a customer complaint is a valid complaint or not. I believe we all know there are often customer complaints that are not valid. For example, how many of you have seen a customer abuse a product and then send it back saying it is nonconforming when it is obvious the product was not installed properly by the customer or otherwise screwed up by the customer?

Every company is different in size, complexity, processes and product(s) - so there is no single answer in how YOUR company should handle these issues, but the above may give you some thoughts to chew on.

from Marc

Typically I ensure clients have a 'Gate' defined in their system. Every nonconformance is entered into a list or database (company size factor), but no request for corrective action is issued until the 'Gate' approves.

That is how I have our system set-up. The problem I am facing is the VP wants to establish more rigid guidelines. Unfortunately, I don't see how this can be done in a way that is automatic without generating a tone of needless CA requests.

Oh well....I will mull it over on the weekend...definitely doing a lot of :frust: , but such is life in the grand world of Quality!

Have a great weekend all!

CarolX

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

You have to set your own requirement. # of pieces, $ amount, etc. Also need to track the ones you do not issue a CAR for, these can actually end up costing you more by the nickel/dime effect.