Is 'Operator Error' as Root Cause ever acceptable?

Wes Bucey said:

Despite the apparent small cost of the error, the real cost is in the soft costs of handling the error and the loss of confidence from a customer who expects perfection.

Click to expand...

The real cost might be in those things, but probably not. It's not good business to go around fixing things that ain't broke in order to avoid phantom costs. It's much more likely that the customer has a system whereby people are not allowed to think about what they're doing, and every NC, no matter how insignificant, requires full-blown CA.

Hello Michael
would be better if you could give us some more details on the steps of the process.
However, if I understand correctly, the operator took the pieces from the BCD side instead of ABC side.

First, I advise you to check the possibility of introducing an error proof system (automatic or semiautomatic, you see) that prevents the operator to take the pieces from BCD side but only from ABC side.
Give us some more details of the process and then we would see how we can help

Ciao
Romanel

Greetings all and thanks for all the great replies.

To answer a few questions...

1. These parts are braid collars that go on the end of a hose assembly prior to welding as support for the fitting attachment.
2. The parts are in bins in the Prep area arranged according to part number which indicates size.
3. The braid collars for this particular mis-pick are in bins right next to each other and the size varies by only .031", but depending on the type of hose of that diameter, will fit on one and not the other.
4. The bins are arranged so that our Prep personnel can pull the correct braid collar for the diameter of hose being prepped.
5. These parts are approximately $1.00 each depending upon the size.
6. Reviewing the Returned Material Log, the last time this error occurred was in April 2005.
7. Management is not going to install or implement any type of bar coding or other stock pulling mechanism due to the low dollar volume of these parts, the infrequency of mis-picks and the costs involved to implement such a system simply are not justified by the error.

So... I've already spent way more time on an isolated incident than is, in my opinion, truly necessary given the situation.

Cheers!!

Michael

I am going to guess this $9.00 part error on your end may have cost the customer a lot more? This alone might drive them to want a CA.
Perhaps your next question would be to find out how this affected the customer other than ruffled the feathers of an inspector or assembler.

Michael T said:

Greetings all and thanks for all the great replies.

To answer a few questions...

1. These parts are braid collars that go on the end of a hose assembly prior to welding as support for the fitting attachment.
2. The parts are in bins in the Prep area arranged according to part number which indicates size.
3. The braid collars for this particular mis-pick are in bins right next to each other and the size varies by only .031", but depending on the type of hose of that diameter, will fit on one and not the other.
4. The bins are arranged so that our Prep personnel can pull the correct braid collar for the diameter of hose being prepped.
5. These parts are approximately $1.00 each depending upon the size.
6. Reviewing the Returned Material Log, the last time this error occurred was in April 2005.
7. Management is not going to install or implement any type of bar coding or other stock pulling mechanism due to the low dollar volume of these parts, the infrequency of mis-picks and the costs involved to implement such a system simply are not justified by the error.

So... I've already spent way more time on an isolated incident than is, in my opinion, truly necessary given the situation.

Cheers!!

Michael

Click to expand...

Just a thought - when I had a customer who required a family of parts, apparently identical, except for an internal depth of a bore which could only be determined by using a depth gage, my own staff rebelled when presented with the engineering drawings, citing their own FMEA (Failure Mode & Evaluation Analysis), which indicated a potential difficulty in keeping the parts separate, since they had to undergo heat treating and plating before being stamped with an indelible ink part number.

The solution we proposed, accepted by the customer, was to add a non-functional groove (up to 5 groves - one for each bore depth) to the outside circumference of the part, making them easily identifiable during heat treating and plating process. Customer eliminated stamping part numbers on each (only used to identify during assembly.)

Net result of mistake proofing: zero mixups, eliminated a two cent per piece stamping charge, eliminated instrument inspection, grooves were essentially free as part of turning operation, we could send more than one part number at a time to heat treat and plating without fear of mixup, saving us soft costs of storage and holding other part numbers while waiting for outside processes and eliminated a drudge job of stamping parts, buying inks, maintaining stamps and stamp pads.

I imagine our customer had similar savings of soft costs, since assembly team only had to differentiate number of rings, not an eight digit part number which could be scraped in handling to become illegible for the significant digit.

Perhaps the parts in question at your operation could be color-coded to differentiate them? Could you ask your supplier?

QC Rick said:

I am going to guess this $9.00 part error on your end may have cost the customer a lot more? This alone might drive them to want a CA.
Perhaps your next question would be to find out how this affected the customer other than ruffled the feathers of an inspector or assembler.

Click to expand...

I like this thinking. Very "customer-centric!"

So we've identified that pickign the wrong part number is a very rare event.

The part number that was supposed to be shipped was a very low cost (and price) part at a low quantity so the cost to you was trivial.

At this point it's a matter of why the customer expects a 'root cause' corrective action and not a simple correction.

As QC Rick pointed out your cost may not reflect the impact to the Customer. (I've had $.05 screws cause tens of thousands of dollars in customer downtime - real cost of paying wages for minimum time worked, not the whole vague "lost opportunity" thing) If the Customer experienced significant cost, then they are probably right to issue a CA and to expect a reasonable corrective action.

On the other hand, the Customer may be blindly issuing CAs for every little incident, and a reasonable logical converstaion will result in them voiding the CA.

Have you talked to your customer about this?

darwinbb said:

I worked as a QA tecnician in an ecoat paint facility in Southwestern Ontario, part of my job is to write corrective action reports whenever customers request for it.
In one of my reports I wrote among others "operator error" as a root cause, our customer rejected the idea and ask me to find some other reason why the problem occurred.
My question is( for those quality people out there) is an "operator error" not a valid root cause in a 7-D format corretive action report?

Click to expand...

Identifying "Operator Error" may be truthful but it isn't very helpful.
I came across a post by Miner from Apr 02 on Causal Analysis where the focus is on Actions and Conditions rather than fault. The idea is that specific conditions must be present for an action to result in an undesirable effect. The undesirable effects are then the conditions for the next actions, etc. A sort of 5-why analysis.
This analysis is then an input, along with other factors (costs, feasibility, etc.) to the decision to effect corrective action(s).
A "do nothing" decision based on costs allows the customer to consider their cost of the nonconformance. If the customer cost is large then a more extensive and expensive corrective would be justified. This, in turn, would justify a price increase for the product.

As a foot note to the original post, kitting parts by assembly and placing them in a tote designed to hold only the right parts in the right quantity eliminates most part pick errors.

See, there is a reason behind the customer rejecting the "operator error " as a root cause. Because, yourself clearly denoting there is a "error" made by the operator,so there is a chance to for the operator to make error.

Then, you have to plug the root causes for the phenomena of the operator making errors. Unless you attached the root cause for the chances of making error, then it will be perennial and may lead to re-occurrence.

The solutions will be : Training (famous and easy step) or you may have to introduce poka yoke "Can Not Make" "Can Not Flow" or "Can Not Recieve" types.

Hope the things are clarified.

Oh! Pick me!

My son totaled his car about three hours ago. He missed a country road turn while driving too fast and tweedling with his stereo or something.

Yeah, he's all right (thank the Maker). And he only did it once. But once was enough. Now he's walking.

1. He knew he should be cautious on these country roads - it was close to home.
2. The weather was perfect and no sun was in his eyes.
3. No animal ran in front of him.

He was just experienced enough to ignore or mentally suspend, just for a couple of seconds, established rules of the road (slow down on those curves, watch the road, both hands on the wheel).

I think mishaps in the workplace happen much the same way: enough familiarity combines with momentary distraction to cause one thing to go wrong in a time and place where everything needs to go right. And, boom. One is doing the 5 Y's.