Situation: Press is set up and run to the point of a 'valid' (correct?) first piece is achieved. We're spooling off of a coil. Many times the operator (Oh, my - Operator Error!) packs parts produced prior to the 1st piece.

Does anyone have any recommendations on how to poka yoke to prevent this? I was thinking a paint or other system which would mark items until the first piece is approved so the setup parts would be highly visible but feel this would be a nightmare.

OK, metal stampers, what say ye?

Two bins?

Marc,

Here's what we do.......

Any parts produced during the set-up of the machine are marked with a yellow paint pen like this

s/u

to identify it as a part produced during set-up. Frequently, we can use these parts for set-up at additional operations (i.e. a formed part from the brake can be used to set-up the spot welder).

I assume the operator is packing set-up pieces, correct?

Not a nightmare....just the KIS system.

Regards,
CarolX

CarolX said:

Marc,

Here's what we do.......

Any parts produced during the set-up of the machine are marked with a yellow paint pen like this

s/u

to identify it as a part produced during set-up. Frequently, we can use these parts for set-up at additional operations (i.e. a formed part from the brake can be used to set-up the spot welder).

I assume the operator is packing set-up pieces, correct?

This is what I'm being told by several managers and a QE there. Believe it or not I suggested this yesterday (day 1) - but was thinking of the machine marking the parts until the setup was complete because I don't trust operators in this setting to mark the pieces.

I'm still investigating the process. Only been there 2 days - but, it's a place I set up the quality system for in 1991-2 so things have changed a lot - but as you know, some things seem to never change. I was looking over the problems and talking to the folks and I keep hearing our favourite phrase Operator Error . The managers are telling me that they go through the setup and (they believe...) that this is what is happening.

That said, it's been a labour intensive effort company for a long time. QS-9000 now and all that rot. I have seen they have taken a lot of hand work out since 1991 and have put in a lot of robots. I sorta feel silly as I sit there and listen to the managers speak about what's going on and you sorta follow, but then I end up as Columbo and finding possible errors in where they're tossing the blame.

I am just starting to review nonconformance trends and such but wanted to start some minds thinking. And... I admit I was a bit hasty in asking - they had just gotten my computer hooked up and I had to go through heck and high water to get internet access but it wwas turned on and I wanted to see that I could access the forums and post without firewall problems.

That said, I'm not a metal stamping expert and although these folks have been doing this for years (metal forming) I want to jump out of their paradigm by hearing from folks here as I investigate problems and try to provide solutions. They have excellent training programs, but, as we all know, sitting at a press all day does not draw the most enthusiastic employees.

Anyway - this is my first 'information fishing trip'. I'm supposed to be at this place most days for a month or two or three (Know whut I mean, Vern?) so it could get interesting. I'm knee deep in Requests for Corrective Actions. Man - it really HURTS to be on this side of the fence. It's darn near work!

I'll provide more details as I get them to try to narrow this down. I'll review the specific process tomorrow and come back with more details. The bottom line to me is it appears most of this will require poka yoke to reduce (I will never say Eliminate - I'm too old and know better...) the possibility of human error.

I am also interested in hearing from others in an effort to benchmark 'Industry Standards' as soon as I have enough parameters to define the processes (they don't do deep draw, for example, and they do a lot of punching with secondary operations like trim after punch. But - I'll start a thread on that soon.<hr>By the way, for those of you who are Harley fans, the place does the stamping, polishing and silk screening for one of their new, fancy bikes - a V Rod or something (I'm really gonna look to see what model that bike is tomorrow). It's a tear drop, polished aluminum alloy with HARLEY-DAVIDSON silk screened on the metal. Goes over where the gas tank would be between ones legs. The bike in the poster I saw there reminds me of the DeLorean - lots of polished metal, only this is an aluminum alloy instead of SS.

No - I can't sneak one out so don't even ask.

Just a question. Why are the parts in the area if they are no good? Seems the quickest solution is to circular file them as soon as you have gleened your information.
Carol brings out a good point though about marking parts to be used for setup in subsequent operations. But other than that, I would say get rid of them before the job is turned over to the operator.

James

We basiclly had the same problem with lathe machined connecting pins --- A solution in that particular work environment was to allow 3 marked setup parts move with the load for use in downstream setup ops.

For others *and often there were a dozen or more *- we isolated them in pans at each station until independant inspectors cold go thru them and determine if they were out of tolerance. I suppose with stampings secondary pallets or tubs would be needed depending on size of the part.

I quieted the management folks claiming OPERATOR ERROR at every turn but it took quite awhile..

Operator error is the same as an "unavoidable accident"? there are no unavboidable accidents - there is no operator error - not as a common noncom cause. Not that it can't happen but in very isolated mis-managed situations.


(((((( So.... requests for corrective action seems like work huh?)
:biglaugh:

Thats one thing I don't miss about working a 6am-5pm job its much easier on the nerves telling someone else they need to do one... :bigwave:

We assign a "pilot lot" work order for new setups. When the parts and process conform to requirements it is released to production. All parts that are produced during the "pilot lot" run are 100% inspected. Those thta conform are placed into production, those that do not are reworked or scrapped.

I worked 18 months, prior to this gig, in a sheet metal stamping outfit who manufactured for Aircraft companies AS9000 wannabees. The operator brought his first piece to the Inspection window. If they chose to run the parts before QC approval, it was at their own risk. Fortunately, they have enough Inspectors so that the wait wasn't too long. What is the scrap rate? Why can't they wait for the OK from QC? Maybe because they were manufacturing for major aircraft companies, they felt the delay to produce something right the first time around was worth the wait. Particularly when dealing with Titanium sheet orders with purchase costs that go through the roof. No scrap sale will recoup that cash. Now, there were times when the Foreman couldn't wait for Inspection, so he made the determination to go ahead. Worked pretty well, most of the time. As the person who processed the rejections, I saw very few times when the product was rejected because 1st piece inspection wasn't accomplished prior to the run. of course, we also had Final Inspection and Customer sampling. The majority of the rejection were for surface finish-handling problems.

And Marc, yes, that's where I experienced all the operator errors as root cause. We, my boss and I, could not convince the company owners that that was not an acceptable root cause. I mean to the point where were chastised for not accepting human error as a cause. There were approx. 100 operators with different skill levels. At least twice a week, we would get an operator error as root cause from a depart ment supervisor. Harldy ever the same person for the same thing. Are you going to foolproof everything? Are you going to train and council everyone because they scrapped (200) $.05 part? It drove me bananas. That's why I'm all ****** up. Be careful, it's catchy. Welcome to the world of presses. Noisy, smelly and very fast paced. :bonk:
:ko: :smokin:

Marc

Is there any other in-process monitoring of the product ?

We operate high speed progression stamping machines, and like you we require a first off inspection to be done, however we also operate Control Charts - hence any problem with the product due to a missed first off would (hopefully) be picked up at subsequent in-process inspections.

Marc said "but was thinking of the machine marking the parts until the setup was complete because I don't trust operators in this setting to mark the pieces" and "They have excellent training programs, but, as we all know, sitting at a press all day does not draw the most enthusiastic employees. "

How does the first good setup part get approved? Who approves it? I'm familiar with a pressing operation where setup pieces cannot be moved to the next step without causing problems. From day 1 all operators are told to throw the setup parts into a scrap container and it has worked pretty well because each setup part must be measured (inspected) by the press operator themselves and they have to record the actual measurement vs. the target on a forming log. I guess the act of measuring the parts and recording the data themselves somehow breaks thru their boredom enough to bring concentration levels up enough to prevent mixing setup parts and good parts. If I were a industrial psychologist (maybe all Q people are a little bit) I’d say you somehow need to make the setup operation “special” – to break the tedium, to show how important it is. That might solve the problem. Whether it is having a supervisor sign-off on the setup (including the scrapping of the bad setup parts) or having the operators themselves do the setup approval and recording of the data with a signature saying the scrap was disposed of, something has to make the setup "special". Hopefully any improvement won't be due to the temporary "Hawthorne effect".


Also, Energy said "that's where I experienced all the operator errors as root cause. We, my boss and I, could not convince the company owners that that was not an acceptable root cause. I mean to the point where were chastised for not accepting human error as a cause. There were approx. 100 operators with different skill levels. At least twice a week, we would get an operator error as root cause from a depart ment supervisor. Harldy ever the same person for the same thing. Are you going to foolproof everything? Are you going to train and council everyone because they scrapped (200) $.05 part? It drove me bananas."

Maybe I'm ****** up too because every now and then I get these crazy flashes in my mind. When I read this post I thought of an error committed by a player on a pro baseball team. Let's say it is one of the best players in the game, very experienced, dedicated, hard working, a regular gold glove recipient as the best-fielding player at that position in the game. Let's say it's Pudge Rodriguez of the Rangers who has won for something like 10 years in a row. His fielding percentage shows he will make an error about once in every hundred opportunities (six-sigma be d**ned). Now, this guy is truly one of the best there is, the top of his profession, yet he makes HUMAN ERRORS sometimes. If you cannot sometimes, in some circumstances, accept "human error" as an acceptable root cause, how would you fill out the corrective action form on ole Pudge?

Just some opinions on a slow day. I welcome input on any of them. :p :thedeal:

Mike S.

Here's another thought I had, but this may not work in your enviroment....works in ours, but we don't do stamping, just punch, form and secondary operations on a much smaller scale and lot size (100 piece orders are average, not a gazillion motorcycle parts).

Our set-up personel are responsible for inspecting their work before turning it in to QC. For the punch press, we maintain an approved parts library for the set-up person to do a visual comparison. This method finds the obvious defects, such as a .250 hole punched at .125. For the press brakes, the set-up person has to inspect identified dimensions to verify the set-up. Once the set-up person has inspected approved the first piece, they can put an operator on it an run the parts. 1st piece is then submitted to QC as a "secondary" check.

In other word, the onus is on the set-up person, not QC.

Have fun, Marc!

CarolX

Mike

I wouldnt request a CA of old Pudge - looks like hes doing fine.

If however he ballsed up every other game then we might initiate a CA.

Point being that you dont raise a CA on every single occurrence, but base it on some longer term trend data.

You would agree I am sure that if he ballsed up every other game it would not simply be human error - he would be considered, for whatever reason, incapable.

I agree 100% on the corrective actions and not requiring one (not to mention a full blown 8-d) for every NC.

This is day 3 so I'll be getting more data soon and better understanding their processes as they are today. I appreciate the comments so far and will provide mre details as I get them.

M Greenaway said:

Point being that you dont raise a CA on every single occurrence, but base it on some longer term trend data.

_______________
M,

Good point, full of common sense, and I agree. But... Sometimes a customer requests a corrective action for a first-time occurrence. Let's say you send some improperly cleaned (dirty) parts to a customer. The cleaning and shipping person just goofed for whatever reason. The customer demands a corrective action be written for them. There is no history of this being a problem at your place in the past, or maybe it happens once a year (>> 99% successful). Same person as always did the job. Nothiong unusual. I'd have to say "human error" was the root cause and it is not economical or practical to implement any changes in the process. Do you shoot me down?

Mike S.

Well, we're way off topic so I'll probably split this thread tonight, but this is a good discussion.

In some instances I can see a 1st time request for CA. But not typically.

A number of factors must be considered. My problem is with companies who request a Corrective Action for each and every nonconformance. Even a 1 piece 'freak failure' item.

The question then becomes - how far do you go and what will they accept? A full blown investigation costs $. In addition, few companies can poka yoke everything. It's one thing if you're producing something for the space shuttle. It's another when we're talking a less critical item.

With respect to Operator Error - we've had a couple of good threads on that here. The bottom line is we all know we can never totally eliminate it any more than we can totally eliminate all machine errors. Stuff happens...

I think the next time I have Operator Error as the cause, I'll change it the effect and change the Root Cause to Operator Partied The Night Before... :thedeal:

Good analogy with the baseball player! Trvth!

We used to use a "dye penetrant" dye to mark the stock before stamping for set up pieces ID...... "

Our actual set up pieces had to be significantly mutilated. The Customer Representatives would rummage through the scrap barrels looking for anything that looked a completed part and would write it up. Unbelievable. We used to keep the scrap room locked and they would ask for the key. Do you think that they had some time on their hands? :eek: :ko: :smokin:

energy said:
The Customer Representatives would rummage through the scrap barrels looking for anything that looked a completed part and would write it up.
____________________
Wow!! I never cease to be amazed about some of the crazy stuff I hear. I've never heard of anything like that. Why the h*ll do they think it was labeled scrap? The day I get nailed for my scrap bucket containing a part that looks good I might just have to take a few days off to cool down else I might do something that would cost me my job. I can only imagine the color of Energy's face on that one!!:mad: :mad:

Mike S.

Mike

With regard to your customers demanding a CA report on an isolated occurrence, I dont think it would be too antagonistic (if your customer understood process variance) to write your report stating that the ocurrence did not warrant any action. Particularly strong if you can prove your process performs at something like a 3PPM defect rate. However if you have not got access to pertinent statistical data just state that this is the first reported occurence, you have audited the process and it appears robust, no further action, will continue to monitor trends.

I have used this approach when necessary and, dare I say 'got away with it'. Many customers purely want a report to file in their own system for their own ISO9001 approval, and pay little regard to the validity of the actual contents. That is not to say you should try and bluff it, if you get the same complaint next month you will probably be in serious trouble with your customer.

So long as you are confident that what has been reported is a true 'isolated incident' then you should be confident in stating that no action is required.

Thats what I think anyway.

I have quite a few years experience in metal stamping quality and I second the 2 bin method. I always mainted a small red painted set up container at the operation prior to set-up approval. After approval this bin was reviewed for any obvious issues and added to the production container. Even with an approved set up part there is a good possibility of a half-hit or mis-hit in the set-up bin so it is good practice to sort the parts prior to adding to the production parts. A great deal of customer rejections are produced during the set-up.

....Our set-up personel are responsible for inspecting their work before turning it in to QC. ........ Once the set-up person has inspected approved the first piece, they can put an operator on it an run the parts. 1st piece is then submitted to QC as a "secondary" check.

In other word, the onus is on the set-up person, not QC.

This should stir the pot...the "lean" guys would say....both set up and QC are Muda....waste.

Are there any stampers out there running without set up or QC personnel?

Just operator self inspection?

It is really hard, lots of training, good standards, methods and accountability are required, but we did it in a foundry, and improved quality and saved money.