Is Poka-Yoke Old Hat? Is Poka-Yoke still a valid concept?

T

The Fixer

#1
Is Poka-Yoke Olda-Hate ??

It's been a while since there was any discussion concerning Poka Yoke and even longer since it's inception. I am wondering if it is still a valid concept or has someone found a better mouse trap ?
Our upper (make that UPPER) management sees our current MIL-STD-105 sampling as an extra step in the process which slows total turn around of the products (printed circuit board repairs). Turn around specs are contained in every customer contract and we are subject to four and five digit penalties if we don't meet quarterly metrics.
Post-Process sampling is, as we all know, rather ineffective for preventing the kind of "inadvertent" mistakes receivers, technicians, labelers, etc typically make. I just don't think human beings are capable of concentrating on ANYTHING for 8 hrs a day.
Poka Yoke seems to be a viable alternative. By the way, we have been ISO9001 registered since 1994, have upgraded to 9001/2000 and are in the process of completing TL9000 upgrades, so QMS is not foreign to this operation.
 

Wes Bucey

Consultant/Advisor
Moderator
#2
The Fixer said:
It's been a while since there was any discussion concerning Poka Yoke and even longer since it's inception. I am wondering if it is still a valid concept or has someone found a better mouse trap ?
Our upper (make that UPPER) management sees our current MIL-STD-105 sampling as an extra step in the process which slows total turn around of the products (printed circuit board repairs). Turn around specs are contained in every customer contract and we are subject to four and five digit penalties if we don't meet quarterly metrics.
Post-Process sampling is, as we all know, rather ineffective for preventing the kind of "inadvertent" mistakes receivers, technicians, labelers, etc typically make. I just don't think human beings are capable of concentrating on ANYTHING for 8 hrs a day.
Poka Yoke seems to be a viable alternative. By the way, we have been ISO9001 registered since 1994, have upgraded to 9001/2000 and are in the process of completing TL9000 upgrades, so QMS is not foreign to this operation.
:topic: For some reason, it has always irritates me when any profession, particularly ours, uses a foreign word or arcane acronym to label a process or technique we expect everyone in our organization to adopt and inculcate into their daily work. Worse, some of our practicioners belittle others who don't "speaka the lingo" as if those folk were somehow less intelligent. Why not use "mistake proofing"? It uses language most people understand. If we used Spanish, perhaps we'd say, "proteja contra errores" [protect against mistakes] (I'm not sure of the exact translation, but I'm sure even my pidgin Spanglish might find a more receptive audience at some of my suppliers than "poka-yoke")
Even widespread usage acronyms like FMEA (failure mode and effect analysis) create stumbling blocks to understanding when the total phrase is used. IMO, we need to use a lot more plain language when we want to get folks to follow our lead.

Back on topic. The phrase "poka-yoke" may go out of style, but the concept of KISS and getting it right the first time were around long before poka-yoke and will probably still be around at some point in a thousand years when no one remembers who Deming, Juran, or Crosby were, either. Why not drop the term and adopt the concept of examining the nonconformities you discover for a real root cause? Then devise a method to avoid the effect of the root cause.
 
C

Craig H.

#3
Wes Bucey said:
:topic: For some reason, it has always irritates me when any profession, particularly ours, uses a foreign word or arcane acronym to label a process or technique we expect everyone in our organization to adopt and inculcate into their daily work. Worse, some of our practicioners belittle others who don't "speaka the lingo" as if those folk were somehow less intelligent. Why not use "mistake proofing"? It uses language most people understand. If we used Spanish, perhaps we'd say, "proteja contra errores" [protect against mistakes] (I'm not sure of the exact translation, but I'm sure even my pidgin Spanglish might find a more receptive audience at some of my suppliers than "poka-yoke")
Even widespread usage acronyms like FMEA (failure mode and effect analysis) create stumbling blocks to understanding when the total phrase is used. IMO, we need to use a lot more plain language when we want to get folks to follow our lead.
[/font]

:topic:

Wes, it may be off topic, but I think this is a very valid point. How can we expect to get "buy in" when we use Martian to speak to Earthlings, our fellow workers?

A rose, by any other name....

Craig
 
R

Rob Nix

#4
Here's a concept: Make consistently good products from stable processes that preclude errors! ;)

If we KAN BAN the use of foreign phrases (kaizen, poka-yoke, jidoka, seiri, seito, etc.), we may be able to convince others we are not as weird as we look; then again, we will take all the mystery out of our important role in business. :rolleyes:
 
T

The Fixer

#5
Hmmmm, I guess I better not use terms like "sigma" either. I didn't invent the term "Poka Yoke" and I had no idea that simply referring to Shingo's work by it's proper name would create such a backlash.
My question was simply meant to solicit recommendations on any newer MISTAKE PROOFING processes or theories that are respected in the quality community. Of course mistake reduction/elimination is a natural outcome of any quality system that we should employ every day. I was not implying otherwise. I am continually impressed by the expertise and insight displayed in this forum. Just thought I could tap the well and get a little advice.
 
#6
The concept of mistake proofing should not be going out of style anywhere that we are trying to deliver quality product to customers. In a recent project, engineering designed a casting that required 4 different length 6-32 screws to attach a PCB to the casting, with a loose lock washer on 3 of the lengths. Use the short screw in the wrong place and the board is not fastened, use the long screw in the wrong place and it punches through the front of the casting. Stray lockwashers can wreak havoc on the PCB. Minor redesign reduced the number to 2 lengths, with integral lockwashers (another change on a molded plastic piece, requiring significant expense, will reduce this number to 1 length. With flow production lines, cross-training, etc., how can can management expect perfection when designers don't involve production from the beginning? Give them more training, tell them to follow the instructions, these things are poor substitutes for a small amount of extra effort on the part of designers to pay attention to parts reductions and simplifications that take very little extra time when implemented in the intial stages of design.

I translate PCB designs from CAD data to our surface mount component placement machine. There are many times that I have choices in where to locate components on the machine. Why would I put two 8 pin IC's, part numbers 8698 and 8968, right next to each other in a feeder? If I did so, am I justified in anger when the operater mixes them up? Is the resulting rework my fault, operator's?
 
#7
Not to worry fixer, I think I can safely say that we know what you mean. It's a different story when the time comes to try to implement things in your company, though. That's when the jargon should be avoided.

/Claes
 
R

Rob Nix

#8
Easy there, Fixer; these comments were tongue in cheek :truce: (and "off topic" :topic: ).

To borrow a phrase, "there is nothing new under the sun". We may freshen up some methods with new names, and may even tweak some methods to further improve their utility, but your experience and creativity in correcting and preventing problems is probably as good as anyone elses here. If you present some specific concern, this forum's panel of experts can then present specific experiences or advice.

I personally agree with you that 'after the fact' sampling plans are never as good as designing out the potential for failure - hence my previous comment's bold and italicized statement. I don't think even Deming was a proponent of sampling plans (e.g. 105-E, F, or whatever).
 
E

energy

#9
No backlash

The Fixer said:
Hmmmm, I guess I better not use terms like "sigma" either. I didn't invent the term "Poka Yoke" and I had no idea that simply referring to Shingo's work by it's proper name would create such a backlash.
My question was simply meant to solicit recommendations on any newer MISTAKE PROOFING processes or theories that are respected in the quality community. Of course mistake reduction/elimination is a natural outcome of any quality system that we should employ every day. I was not implying otherwise. I am continually impressed by the expertise and insight displayed in this forum. Just thought I could tap the well and get a little advice.
I don't see backlash. You started a good thread and it has resulted in a very good discussion. There are too many "special" names for things we do all the time. There is also special interest to keeping them that way. I mean, in order to sell it you have to convince others that there is always something better. You have tapped the well and got, I feel, good responses. As for sick "sigma", there's plenty of discussion here on that. Why, it even has its own forum. Good thread starter. :agree:
 
S

Simon Timperley

#10
I mean Wes you practice Kung Fu (punching and kicking) on the poor guys head cause he says "Poka Yoke" on a Quality Discussion Forum…and the rest of you back slap and applaud him and then there's this mighty cover-up / u-turn.

Remember Poka Yoke wasn't a foreign expression to Shigeo Shingo (the Japanese guy who invented the concept) and maybe when The Fixer is not amongst learned QP's it is possible he refers to Poka Yoke as failsafing, mistake proofing, or even Chumba Wumba as I like to call it.

I don't think it's the terminology that's the problem IMO it's more the wazzocks who show and tell the tools either arrogantly, incoherently or both.

Poka Yoke is not about examining the root cause of nonconformities and taking preventive action as you advise Wes, it's about developing tools, techniques and processes such that it is impossible or very difficult for people to make mistakes in the first place (prevention). Or where it is not possible to build in failsafes a poka yoke could be a device that signals the user when a mistake has been made, so that the user can quickly correct the problem (detection).

Well worn examples:

Prevention device - you can't start a microwave oven with the door open
Detection device - car beeps if I leave the lights on

Until robots can wipe our butts effectively there will always be the need for human interventions in the workplace and consequently there will always be human error. IMO Poka Yoke devices are as valid now as ever and can be cost-effective and simple methods of preventing defects, saving money and sometimes lives.

There are some great examples of Poka Yoke devices on John Grouts page here:

http://www.mistakeproofing.com

Long live Chumba Wumba!

Regards,
Simon
 

Top Bottom