Bloomberg reports that Toyota's vision for the future includes replacing (some of) its robots with humans.
While it may sound nonsensical, the image starts to make sense when considering some aphorisms on automation (for example, "Computers have enabled people to make more mistakes faster than almost any invention in history, with the possible exception of tequila and hand guns") and the recent recalls of millions of cars.
It's far from suggesting that robots were causing the faults or that, by replacing robots by humans, the company will diminish the rate of reproducing the faults with high efficiency. However Toyota has had a culture for quality for quite a long time and seems to have identified a cause for the lack of quality (which caused the recalls) in the distance the computers and automation place between the workers and the product they manufacture.
Learning how to make car parts from scratch gives younger workers insights they otherwise wouldn't get from picking parts from bins and conveyor belts, or pressing buttons on machines. In an area [...] at the forging division of Toyota's Honsha plant, workers twist, turn and hammer metal into crankshafts instead of using the typically automated process. Experiences there have led to innovations in reducing levels of scrap by 10% and shortening the production line 96% from its length three years ago. 760 workers take part in 96 percent of the production process at its Motomachi plant in Japan Toyota introduced multiple lines dedicated to manual labor in each of Toyota's factories in its home country
"We cannot simply depend on the machines that only repeat the same task over and over again," Kawai said. "To be the master of the machine, you have to have the knowledge and the skills to teach the machine." Kawai, 65, started with Toyota during the era of Taiichi Ohno, the father of the Toyota Production System envied by the auto industry for decades with its combination of efficiency and quality. That means Kawai has been living most of his life adhering to principles of kaizen, or continuous improvement, and monozukuri, which translates to the art of making things.
(Score: -1, Troll) by Anonymous Coward on Tuesday April 15 2014, @01:12PM
(Score: 2, Insightful) by Anonymous Coward on Tuesday April 15 2014, @01:54PM
What editor?
I don't want to be a grammar Nazi, but this is pretty basic stuff.
(Score: 3, Insightful) by mattie_p on Tuesday April 15 2014, @02:52PM
Editors are human too, and even though we have great built-in error detection mechanisms, sometimes those break down. Thanks for checking us, I've made some corrections. Thanks for reading.
~mattie_p
(Score: -1) by Anonymous Coward on Tuesday April 15 2014, @01:55PM
(Score: 3) by c0lo on Tuesday April 15 2014, @02:00PM
If you don't like it, try "To err is human, but to really foul things up you need a computer."
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: -1, Offtopic) by Anonymous Coward on Tuesday April 15 2014, @02:21PM
(Score: 2) by c0lo on Tuesday April 15 2014, @02:30PM
Was it? Can you elaborate why do you think so? Maybe explain a bit what was the reference to gun politics about?
(does the quote not support the point of: "Have a faulty design and use automation to reproduce it with high efficiency may not be a path to success"?)
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2, Funny) by Anonymous Coward on Tuesday April 15 2014, @01:18PM
dey took our jerbs!
(Score: 2) by RamiK on Tuesday April 15 2014, @01:45PM
If Toyota can afford manual labor instead of automation, it probably means minimum wage is roughly the equivalent of nothing more then bread and water.
compiling...
(Score: 3, Insightful) by threedigits on Tuesday April 15 2014, @01:49PM
Toyota does plenty of manual labor. it has more than 300,000 employees, after all. The key is doing it where it matters.
(Score: 3, Interesting) by VLM on Tuesday April 15 2014, @03:00PM
Another way to look at it is very little direct labor is embedded in the cost of a car, so experimenting to improve quality and reduce costs, isn't very expensive in the big picture.
I can verify that the Venn diagram shows massive overlap of the best manual machine operators and the best CNC machine programmers and this is not seen as even remotely controversial in the field. And neither groups works for bread and water (LOL).
Fundamentally the whole program seems oriented around thinking really hard about how to improve processes based on experience, pretty typical industrial engineering although a somewhat unusual way to implement it, and those guys don't work for bread and water either.
(Score: 3, Informative) by frojack on Tuesday April 15 2014, @06:24PM
Quoting Summary:
Point One: That is never going to work.
Point Two: Even at minimum wage, the chance of a crank shaft coming in cheaper when built by the methods they describe is about 1 in 100,000. Which is also the likelihood of achieving a working crank shaft.
That line in the story says to me the author is an idiot, and didn't understand what they were looking at, or more likely they never toured the plant at all.
No, you are mistaken. I've always had this sig.
(Score: 1) by mrchew1982 on Tuesday April 15 2014, @08:21PM
I believe that the intent of the author was to illustrate some of the tasks that are performed by humans *in order to learn* how to program the machines to do them more efficiently. There is no way that Toyota could be cost competitive if they used human labor in most parts of auto manufacture, that just not how it's done anymore. I see this as a hands on learning experience to give their process engineers an understanding of what the heck they're programming robots to do. No surprise that it's led to improvement!
The most prized designers (architects, engineers, etc) are those that have a firm grasp on reality and not some pie-in-the-sky book knowledge that they were taught in school.
Ditto for CEO's and managers, you have no business managing people in something that you are not competent in yourself!
(Score: 5, Insightful) by Covalent on Tuesday April 15 2014, @01:48PM
This sounds like just good quality control and employee education. The end of the summary suggests that Toyota can't rely on robots that just do the same action over and over. It does not, however, suggest that more advanced robots are a bad idea.
Any robot that does not have a robust error checking mechanism should use the best one available: A pair of human eyes.
You can't rationally argue somebody out of a position they didn't rationally get into.
(Score: 5, Insightful) by c0lo on Tuesday April 15 2014, @02:12PM
A bit more than that, in my understanding: Toyota also involves employees with good practical experience to operate the machines manually and discover, by direct experience, opportunities to improve the efficiency in both processing time or amount of material used (or scrapped). Of course, once discovered, they'll "teach" the robots better
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2, Insightful) by LoRdTAW on Tuesday April 15 2014, @02:58PM
Sounds like they are creating skilled labour. Skilled labour normally commands a higher salary which I hope Toyota pays. When people (managers/bosses) hear about automation it is seen as a tool to move a skill from a worker to a machine. In some cases it works and improves production and quality. But all too often it is seen as a way to eliminate skilled workers and replace them with machines staffed by disposable minimum wage drones.
The reality is: If you pay a decent wage you have better employees who produce more and care about the quality of their work.
(Score: 1) by CoolHand on Tuesday April 15 2014, @02:23PM
No, it is way more than that, and very insightful, IMHO. It is actually saying that without good hands-on experience, enginners and QC people can not have full intimate knowledge of the materials in order to assure quality and have the understanding to make further process and material improvements. This is the sort of thing that can only truly be possible with robotics when AI becomes so advanced that is is truly artificial intelligence of at least a level with humans. Possibly not even then, unless a method can also be determined to impart a human's lifelong sensory perceptions into the AI/robot.
Anyone who is capable of getting themselves made President should on no account be allowed to do the job-Douglas Adams
(Score: 2) by c0lo on Tuesday April 15 2014, @02:38PM
A strong AI won't emerge without a good amount of sensory perception. Even more, it may be that the AI senses will need to overlap quite a lot over human senses for a human/AI communication to make... well... sense.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 0) by Anonymous Coward on Tuesday April 15 2014, @04:44PM
And even if there were a strong AI, it would not be a given that it could completely replace humans in this area. After all, what does an AI know about how humans perceive a car? For an AI, human perception and human feelings would always be alien. It could reason about them the way we can reason about the bee's ability to see UV. But would you think you could design a flower that's attractive to bees, other than by trial and error?
Something that an AI might consider an irrelevant irregularity might be seen by a human as major flaw. On the other hand, and AI might spend considerable time solving a cosmetic "problem" in the production process that humans wouldn't even notice.
So, to exaggerate, you might get pre-bumped cars (because the AI considers the bumps caused in the production process irrelevant) with a colour uniformity at the limit of human perception (because the AI considers changes in the colour which we would hardly notice as serious).
(Score: 5, Informative) by bucc5062 on Tuesday April 15 2014, @02:01PM
"We cannot simply depend on the machines that only repeat the same task over and over again,"
That is too rich. Once again humanity has to shoot its self in the foot to discover that it is not good to shoot your foot for it really hurts and it makes one less productive. What's next, the banking industry realized that consolidating all the wealth, capital, and liquidity in a small group could be bad and that moving money around is much better (but only after one more crash, got to really get the point). How about politicians who wake up one day and go "oh my god, being a rancorous, obstructionist, idealog representative is hurting the country I was sworn to defend, I better start working with my fellow members to make life better for everyone".....
ho ho ho ha ha ha ha
For some reason this article reminds me of this Asimov story, the feeling of power [themathlab.com]. Enjoy.
The more things change, the more they look the same
(Score: 1) by urza9814 on Tuesday April 15 2014, @11:16PM
Uhh, what?
The point of the article is that they can't depend on these machines alone to improve their process. How, exactly, is a robot that always makes the same cuts in the same way going to figure out that it can better optimize to reduce scrap material?
It won't. That's why you need humans. But a CAD designer sitting in some office tens or even thousands of miles away from the factory could easily miss these improvements too.
They're having a small number of people produce a small number of parts so that those people will know how those parts are actually produced. Seems like a pretty good idea to me. Toyota isn't the first company to figure this out either. GE did it a couple years ago:
http://www.theatlantic.com/magazine/archive/2012/1 2/the-insourcing-boom/309166/ [theatlantic.com]
In that article, it was assembly line workers in China rather than machines, but the point is the same -- design and manufacturing work a lot better when there's communication between the two. And you obviously can't communicate much with a robot, which explains why Toyota is bringing back some small amount of human labor.
(Score: 5, Interesting) by Geezer on Tuesday April 15 2014, @03:57PM
I was an automation specialist at Toyota's Aisin Warner transmission plant in N. Carolina for years, and it was quite common in the Kaizen process to continuously evaluate every manufacturing step for optimum value. Value sometimes means asking, "Is a $1.5M automation solution really desirable over a $16.00/hour human?" I can recall several times where the automation was so expensive to keep running we pulled the cell and let a human do it. It always comes down to efficiency, and Toyota engineers are expected to exercise sound business judgement. That means evaluating every task and selecting the optimal solution, be it robotic, human, or hamsters on a tread mill.