Has anyone ever addressed a customer corrective action with no corrective action needed? Is this acceptable. I work for a distributor and our customer is rejecting one lot of parts because one nut failed torque test requirements. Lot size is 650 pcs. and they tested two nuts with one failing. Mfg. inspection reports shows 10 pcs. tested passed requirements. This is a large OEM so I do not want to **** them off but finding root cause of one nut non-conformance is difficult if not impossible.

Gary,
Are you the manufacturer? If you are a distributor, you may have to ask the manufacturer to increase his sampling plan. A torque failure is a serious problem. You have evidence of 0 in 10. The Customer has evidence of 1 failure in 2. Did your Customer do them all? Tell your customer that the manufacturer has agreed to increase the amount of torque tests per lot. But, make sure that they do it. Or, look for another Supplier. It may be a swaging, brazing or clinching process defect. If your Customer has signed off on your Quality Plan, you may suggest that the product met current requirements, but you will increase the testing by (50%?) to ensure that there are no failures. Personally, I would insist a few lots at 100% to convince myself that it was a rare occurrence. It's a problem that requires Corrective Action. Your statement that it is not, falls on these deaf ears!:smokin:

Energy, The customer did not test them all. The Mfg. tested 10 pcs. as did I as soon as I was informed of the reported condition with the same lot of parts pulled from stock. Could not duplicate condition. Since 1998 we have purchased 70,000 pcs. comprised of 19 different lots with no reported failures. I do not see a trend here and would like to call it a isolated incident. The one nut failed on the seventh removal cycle.

Gary,

Then you explain just like you did here. I copied this reply from a post here several months ago, for potential use by me if this sort of condition arose. See if you can use it:

"Current review of the system is sufficient to catch the vast majority of errors. The rest of the Quality System has been audited and found to conform to requirements of ???? Further preventive actions to eliminate the cause of potential future occurrences(Check spelling) would not be commensurate with the minimal risk involved."
All this after you explain the history of failures as you did in your post. Give it a shot and good luck!:bigwave: :smokin:

I have used the "in all of our data..." approach a couple of times. If it is truely a 1 in a million case so to speak, there isn't a lot of justification to spending tons of money to fix a problem that isn't there. One thing you need to be aware of, if this is the route you go, you have to sell this theory to your customer. They aren't going to just take it lying down in most cases. They want to see you do something that makes them feel warm and fuzzy.

Overall, back to the original question, yes there can be customer issues that do not require corrective actions. Not real often. There is usually something you can do to either make you or them more confident of the process. But I can remember a couple of stand-out issues from the past. One, the customer picks up their material, drive off down the road, falls asleep and wrecks their vehicle then turns in a claim that our product surface is bad, dinged up, bent, scratched, or whatever. Ya think? Lucky for us, we had instituted a new preventive action program by taking a polaroid (upgraded to digital) of each truckload before it left the shipping bay.

SteelMaiden,

We also photo everything we ship to preclude Customers and Freight Companies from saying that "This is how we received it". It has saved us from additonal expenses that were not our problem. Here's another example of a Customer Corrective Action Request that was enjoyable. During an audit at our facility, the auditor issued a non conformance because we didn't have the latest revision of a particular packaging procedure. To begin with, they never sent it to us. Second, we never received an order with this procedure specified. When we responded, he just sent us the latest spec and never retracted the N/C. It remains in their "Historical" data.

Same Customer. Late shipments because "their" designated carrier never picked up the shipment that was ready. Our complaints go unanswered and we get their monthly Supplier Status Report with the late shipments affecting our rating. But, guess what? They keep on purchasing and remain one of our largest Customers. Sometimes, lower echelon people just like to thump their chests and try to intimidate you. If these problems, Customer generated, caused us to lose their business, you can bet our CEO would take "our" history to their top person and it would get rectified. Possibly resulting in some "counselling" for the real responsible people. You never rule out what may happen when the "Big Suits" get together.
O.K., this was longer than I intended, but my passion for resisiting the "bully" type customer gets me cranky. Later "Maiden":smokin:

Originally posted by energy
SteelMaiden,

... When we responded, he just sent us the latest spec and never retracted the N/C. It remains in their "Historical" data.



Around here we refer to that as "The Hysterical Data File" All the stuff that people think they can get away with.

Gary,

Are we digging too deep here? There had to be a reason for the nonconformance.

Are you within the PPM requirements for the customer?

Was a new person running the line?

Have you looked at maintenance and set-up records?

What does your FMEA define?

Are increased audit rates applicable until the root cause is determined?

I guess I am getting to the point of let's get real. Something malfunctioned in your system. It may be training, it may be execution of instruction, it might be a lazy employee.

Whatever the scenario there is a glitch somewhere in the process. I surely can't see or say what it is or whether the person that wrote the C/A was just having a slow day.

What type of traceability do you have?

You could not reproduce the incident, did you request the defective part from the customer so you could perform your own test on the subject part?

Have you visited the customer to determine if the operator had the air ratchet backwards and was afraid to open him/herself to superior review?

There are just so many variables that it is a very hard question to answer.

Please believe me that this is not a dressing down, just an opinion of an old *art.

Thanks for all your replys. Keep in mind that we are a distributor not the manufacturer. We purchased the parts from the largest nut manufacturer in the world, Simmonds, located in france. We have received the nut back that is in question and the rest of the same lot we sent them. We tested the nut and agree with the customer that torque effectivity on the seventh removal cycle is below specs. We plan to send the lot back to the manufacturer in France for verification of the rejection. The manufacturer tested ten pieces at time of manufacturing and test reports reflect product passed all torque tests. We tested ten pieces after we were informed of the reported condition. We could not duplicate the condition. I feel our customers acceptance on 0 defects and reject on 1 is ok but what is the fallout of the lot. Is there not such a thing as varibles in the manufacturing process that may allow one bad nut to go undetected due to the inspection sample size at time of manufacturing? How many pieces should be tested before you get a warm fuzzy feeling about the entire lot. I think ten pieces on a 660 piece lot size is appropriate. I can understand implimenting some form of corrective action if I could find more than one nut not to spec.

Gary,

You are in a tough position, what was the root cause and fix of the corrective action you issued to your supplier?

Al, I have not yet received a response on the corrective action I issued to the manufacturer. I have been informed that a response can take up to two months when you are dealing with France, but my customer wants me to address his corrective action in 10 days. This is a giant manufacturer, and we are a small distributor so it is very difficult to ask them to change their investigative procedures to address the issue quickly.

Caught between a rock and a hard place, get confirmation from your customer as to how long a C/A will take and wotk with your customer to determine the next step. If the customer is large enough they may be able to expedite matters.

I think they will still need some respnse from you as to how you are going to "inspect" the suppliers product until a solution is developed.

On a personal side I will say that the person that issued the corrective action must have been having a slow day and needed something to do.

How often do you visit the customer ?

Al, I guess we dont visit them enough and this is their way of getting to know us better! I will keep you posted on the outcome

I didn't want to start a new thread because the title to this one seems appropriate enough...;)

How do ya'll close out your CARs? Does the originator do the initial verification of the Quality Action Plan and say that it is effective and sign off on the CAR and send it through to Quality? Or does your Quality dept do all the verifications? And if so, do you wait until the next internal audit or do you do special CAR closeouts?

What I thought we might do is to have the originator of the CAR be the one to follow up on the effectiveness and allow the CAR to be closed, and then whenever that dept was audited on internal audit, the auditor would perform secondary verification of the closed CAR. Is this the right way?

Lucinda,
Our plan is to assign an auditor to audit all previous audit findings before the next system audit. Currently, that would be no later than 30 days. Obviously, if it was a serious violation (I don't believe in classifications), prompt action should be taken. YOU determine and implement the corrective action needed.

Incidentally, I have printed the 6 procedures you cited another post and hung it on the wall where I will study it until I see what you see!:bonk: :smokin:

Originally posted by energy

Incidentally, I have printed the 6 procedures you cited another post and hung it on the wall where I will study it until I see what you see!:bonk: :smokin:Ya lazy :ca: ! Find the thread and post a link for us so we know what yer talkin' 'bout! :thedeal:

By the way, yer spellin' is sloppy, too! ;)

http://Elsmar.com/Forums/showthread.php?t=3889

Admin: I'm not a lazy :ca: I just don't know how to provide a link in your hotshot ISO Site :) :ko: :smokin: And I have a cold that affects my spelling :) :smokin:

Lucinda, we have focused on this recently.

In our revamped process, we do make the originator responsible for not only accepting the corrective action plan, but following up with the person responsible to ensure it is done and in a timely manner too! We have established guidelines for the quality impact of the RCA (which is assigned by the QA Manager or Engineer) that have specific associated time requirements for response and implementation, as well as follow-up.

This is tracked by me (open RCAs), reported weekly to all involved with any open RCAs and both the QA Manager and Plant Manager.

The improved system has resolved a number of our process problems and we have sustained the process. So, I am quite pleased since it was my baby to put to bed! :ko: (so to speak, of course). Hope this helps.

Maybe I don't have a clear picture of the situation but it appears to me that the evidence suggests you probably have a real problem that you are trying to deny.
Is this a go / no go test? If it is than your sample size is way to small for any reasonable attribute acceptance sampling scheme.
If you are doing variables acceptance sampling, what are the M values calculated etc.
Or maybe you are doing more progressive control charting. Is it out of control?
You have taken a total sample of 12 from a lot of 660. Ten at the manufacturer and 2 at the customers. You have a confirmed failure of one. The 95% confidence interval for the population is from 0.2% in the lot to 38% defective on such a small sample.
You say you would do corrective action if more than one were found. This implies you want better than 0.3% quality.
If you follow ANSI/ASQ Z-1.4 guidelines for lot size of 660 and an AQL of .4% you have already failed in that it would call for code letter J and 32 pieces with Ac-Re 0-1. So the implication is that your quality level is less than 0.4%, i.e. more than 0.4% defectives are in the lot. How many more than one is not known.
Why don't you test enough to narrow the confidence interval and reassure the customer ? You may be right in thinking this was a one in a million incidence, but your case would certainly be stronger with data rather than conjecture.

I was reading back through some old threads 'enhancing' some thread titles and checking for deal links and I came across this oldie but goodie. Just thought I'd bring it up for added discussion.

Have you replied to a customer with No Corrective Action Needed? What was the customer's response?

I was reading back through some old threads 'enhancing' some thread titles and checking for deal links and I came across this oldie but goodie. Just thought I'd bring it up for added discussion.

Have you replied to a customer with No Corrective Action Needed? What was the customer's response?In my later years, I became very political in dealing with CARs from customers.
It seems to me that a large percentage of customer-initiated CAR (as many as 20%) were due to customer error in inspection or handling/installation. An equal percentage were just silly (one-time orders where the nonconformance was signed off by customer engineers, but some form-happy clerk wanted a CAR.) A small percentage of CARs (1% - 2%) we NEVER satisfactorily determined the root cause and failed to detect any other examples of nonconformance.

In every instance, regardless of the merit of the CAR, we documented a ritual of investigation:

Confirm the nonconformance with our own inspection
(if we couldn't confirm, we would conference with customer, sometimes involving 3rd party inspection services, until we confirmed or until customer acknowledged his error.) If customer error, process stopped.
Determine if root cause investigation were economically feasible (often it was not for one-time custom orders.) If not, offer replacement or refund and stop process.
Conduct root cause investigfation when economically feasible.
Determine if corrective action is feasible and within our capacity and capability. (If not, this was reported to customer and process stopped.)
Perform corrective action, evaluate, and report to customer.
It is important to note we NEVER broke communication with the customer throughout the process. When I am the customer, the worst thing a supplier can do is leave me in the dark about the progress of a CAR. I have more respect for the guy who tells me to drop dead than the guy who tells me nothing.

I was looking over some old post while I'm sitting in the hotel room and came across this one.

What's interesting is the customer mentioned in the original post is who I work for. While I did not originate the SCAR, my group did. I enjoyed the comments presented here. I don't remember the final outcome of the corrective action as it's been too long and many SCAR's ago, but I believe the inspection was increased at the distributor.

Prior to sending the SCAR to a supplier, we investigate the defect and have a team meeting to discuss all the possibilities. We evaluate whether a written response is required or not depending to the circumstances. Unfortunately, our suppliers may not know what we are seeing as problems in our business and what our customer is telling us.

The reason we do 10 day cycles is because we can't afford to have a catastrophic failure during a mission. If there is a problem, we need to know quickly. A bad nut will destroy a missile just a easily as a major sub-system, not to mention the potential problems for the war fighter.

In the previous post by Wes Bucey, "It is important to note we NEVER broke communication with the customer throughout the process. When I am the customer, the worst thing a supplier can do is leave me in the dark about the progress of a CAR. I have more respect for the guy who tells me to drop dead than the guy who tells me nothing." As a customer, that is a continuing problem for us. When we issue the SCAR, we talk to the supplier about the defect, possible corrective action and the due date. Then we call prior to the due date to make sure the supplier is working on the response. If the supplier misses the date, they are subject to being removed from the Approved Suppliers List until it's resolved. That action can be prevented with communication but it should come from the supplier.