Boeing Fires Its Fuselage-Assembling Robots, Goes Back To Using Humans (seattletimes.com) 100
schwit1 quotes the Seattle Times:
After enduring a manufacturing mess that spanned six years and cost millions of dollars as it implemented a large-scale robotic system for automated assembly of the 777 fuselage, Boeing has abandoned the robots and will go back to relying more on its human machinists...
The technology was implemented gradually from 2015 inside a new building on the Everett site. But right from the start, the robots proved painful to set up and error-prone, producing damaged fuselages and others that were incompletely assembled and had to be finished by hand. "The Fuselage Automated Upright Build process is a horrible failure," one mechanic told The Seattle Times in 2016. Another called the system "a nightmare" that was snarling 777 production. Yet Boeing insisted then that these were teething pains that would pass... The automation has never delivered its promise of reduced hand labor and Boeing has had to maintain a substantial workforce of mechanics to finish the work of the robots. Because of the errors in the automation, that often took longer than if they had done it all by hand from the start...
It's taken six years to finally throw in the towel.
Yet the article also notes that Boeing will continue to use its highly-automated autonomous robotic systems on other parts of their 777 assembly process.
The technology was implemented gradually from 2015 inside a new building on the Everett site. But right from the start, the robots proved painful to set up and error-prone, producing damaged fuselages and others that were incompletely assembled and had to be finished by hand. "The Fuselage Automated Upright Build process is a horrible failure," one mechanic told The Seattle Times in 2016. Another called the system "a nightmare" that was snarling 777 production. Yet Boeing insisted then that these were teething pains that would pass... The automation has never delivered its promise of reduced hand labor and Boeing has had to maintain a substantial workforce of mechanics to finish the work of the robots. Because of the errors in the automation, that often took longer than if they had done it all by hand from the start...
It's taken six years to finally throw in the towel.
Yet the article also notes that Boeing will continue to use its highly-automated autonomous robotic systems on other parts of their 777 assembly process.
That sucks (Score:1)
I trust robots to do a better job than human workers.
Re: (Score:1)
Re: (Score:2)
Robots will never replace humans for humans can be mass produced by unskilled labor.
Re: (Score:2)
+1!
(And many robots are actually assembled by robots. It is when they are programed by robots that we are doomed.)
Re: (Score:2)
But humans are expensive to operate.
Re: (Score:2)
But humans are expensive to operate.
Only if you pay them.
Re: (Score:2)
Pay them? Doh!
Re: (Score:2)
If you call 9 months of complex biological processes followed by 18 years of instruction and socialization followed by more training and socialization unskiilled.
Re: (Score:1)
Re: (Score:2)
Of course there's another way of looking at it. The "complex biological process" is entirely automated and requires no skill. Cockroaches have children. Voles have children. Chickens have children. How skilled are they? That argument reminds me of the famous "I'm a MOTHER" line - as if that was some sort of special qualification.
Then you have your 18 years of "instruction". It's not 18 years of instruction. It takes 18 years for us to trust them to reach a basic minimal level of competence. The instructio
Even voters have children (Score:2)
At first I misread that as:
"Cockroaches have children. Chickens have children. Voters have children. How skilled are they?"
Cockroaches aren't brilliant, chickens are kinda dumb, and the voters are a special kind of stupid. :)
Re: That sucks (Score:2)
Re: (Score:2)
Without reobotic intervention, the LGTQs cannot replicate.
Re: (Score:3)
Robots will never replace humans for humans can be mass produced by unskilled labor.
Unfortunately, the quality of the humans varies extremely.
If fact, a huge percent are not very useful for anything at all.
Re: (Score:3)
Re: (Score:2)
There's need and there's people who say they have need, and you will bankrupt yourself going case by case to catch all the fraudsters.
That's why simplifying systems works better than chasing down offenders. Take UBI for instance, if you give it to everyone then the only fraud prevention you need to do is making sure that everyone who gets it exists.
Re: (Score:2)
squatting on someone else's property for years in the hope of exploiting the law and grabbing that piece of land.
So naturally, finding such behavior reprehensible, you have looked up the title holder's address in county records and sent a warning letter so he/she can issue an eviction notice and bring the squatter's claim to a halt, right?
Re: That sucks (Score:4, Funny)
If fact, a huge percent are not very useful for anything at all.
They keep Slashdot in business. ;-)
Re: (Score:2)
Re: That sucks (Score:2)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
Re: That sucks (Score:2)
Don't feel bad; most places that just makes you a normal person. On here, however, you'll likely find yourself unable to wield any "retard strength."
Re: That sucks (Score:2)
Re: (Score:2)
Evidence suggests that trust is misplaced.
Yang will be disappointed (Score:2)
Automation doesn’t always lead to the creation of a post-labor society? Who knew.
Re:Yang will be disappointed (Score:5, Insightful)
The technology is not mature at this time, also because the whole rest of the process is not really geared towards it. Eventually, this will work, but currently people try to do this on the cheap (because "robots") and cut corners that cannot be cut without compromising the whole thing. Give it another 20 to 50 years.
Re: (Score:3)
It's not that technology is "not mature". It's that it's "more expensive".
Human body is an amazing biological machine. Just because a robot can do certain very specific tasks better than a human body, it doesn't mean that it can do everything it can do, most things that it can do, or even a meaningful minority of things it can do for less resource investment than what it would require for a human body to do it.
This is the case where they probably could ultimately make it work. It's just that such a robot wo
Re: (Score:3)
I disagree. Robotics is in its infancy. It _will_ replace humans in basically all mass-production jobs. It will just take quite a bit more time for it to get there.
Re: (Score:3)
I disagree. Robotics is not at all in its infancy. It's very mature in many industries. It's in its infancy in large-scale precision production.
For applications like the aerospace industry (and I guess I'd lump Tesla in there too) where the scale of work is very large and/or tolerances are very low, it doesn't seem to be quite mature yet. But if you look at a canning factory, robots can and do fill and seal thousands of cans of foodstuffs every minute. There the scale is small and the tolerances aren't that
Re: (Score:2)
Robotics is not at all in its infancy. It's very mature in many industries. It's in its infancy in large-scale precision production.
So what you're saying is that it hasn't developed its fine motor skills yet? That sounds like infancy to me.
For applications like the aerospace industry (and I guess I'd lump Tesla in there too) where the scale of work is very large and/or tolerances are very low, it doesn't seem to be quite mature yet. But if you look at a canning factory, robots can and do fill and seal thousands of cans of foodstuffs every minute.
So the difference is that if it's a very simple job then the robots can do it, but if it's complicated they can't, because that level of feedback hasn't been developed yet? Yeah, that doesn't actually sound mature.
Re: (Score:2)
Indeed. As replacement for classical inflexible mass-production machinery, they may do well in some spaces, but the real core skill of robots is flexibility as they can adjust to anything on-the-fly (hardware wise). The software cannot really do that yet, sensors are lacking or not good enough, integration with other processes needed for that flexibility is lacking, design tools for the good produced that can deal with this are lacking. Hence "infancy".
Re: (Score:3)
It's not in its infancy in large scale precision production either. That's one field where robots are clearly superior to humans.
The problem is that robots excel in precision in fully pre-programmed tasks that require minimal on the fly modifications. Preferably none. Even minute modifications rapidly build up required complexity in robots.
And when you're working on something bigger than "machine this simple part to extremely tight tolerances", you typically must start modifying the process on the fly for e
Re: (Score:2)
Unsubstantiated assumption you persist in making: "robots can replace human body in an economical fashion".
Because it's not enough to be able to do what humans do. They must do it more economically than a human body. Robotics are a very mature field, which is why we can make amazing robots that could replace humans in most things today.
The reason why we don't? Such robots are astronomically expensive to build and maintain, far eclipsing costs associated with training and paying humans to do the job.
This is
Re: (Score:2)
Sorry, but robotics is _not_ a mature field by almost all indicators that need to be taken into account when making such a judgement.
Re: (Score:2)
There's a good rule that when someone makes a specific argument, and refuses to name any specifics, they usually know they're wrong and just don't want to admit it.
In your case, that's "it's not a mature field by almost all indicators [none of which you will provide as an example]".
Re: (Score:2)
So I should compensate for your laziness? Sorry, but no. Requiting the other side to provide examples, but not providing any evidence yourself is is just dishonest.
Re: (Score:2)
So not only are you going to double down, but you're also going to accuse me of my sins, which I am innocent of. As I have indeed provided at least one example in this thread.
I guess we know why you're so bent on insisting robots are so capable. You see yourself as an intellectual analogue of a robot.
Re: (Score:2)
I guess we know why you're so bent on insisting robots are so capable. You see yourself as an intellectual analogue of a robot.
Excuse me? I did state the opposite!
Re: (Score:3)
Automation doesnâ(TM)t always lead to the creation of a post-labor society? Who knew.
We've always found things for people to do after automation in the past, and I expect that to continue.
One thing that happened with automation was the ability to have the time to learn different trades. The use of factories meant fewer people were needed to produce clothing. This meant people could afford more than a single pair of pants and a single pair of shoes. What this also did was mean people could take the time to become entertainers, school instructors, and far more. This meant the ability to r
Re: (Score:3)
Automation doesn't always lead to the creation of a post-labor society? Who knew.
We've always found things for people to do after automation in the past, and I expect that to continue.
For example, after buying your spouse a vibrator, you'll have more free time to read, etc...
Re: (Score:2)
For example, after buying your spouse a vibrator, you'll have more free time to read, etc...
If you can be replaced by a cheap piece of plastic maybe you should be...
Re: (Score:2)
Re: (Score:3)
"We've always found things for people to do after automation in the past, and I expect that to continue."
The next generation (other people), yes. The same people, no.
Re: (Score:2)
We've always found things for people to do after automation
We've never had full automation before. We've had increases in production and efficiency, yes. We've automated large parts of the process, yes. But we're getting close to the point where a factory will only require a manager to push the "on" button and a small maintenance crew to keep the robots running. Everything else will be fully automated, from receiving raw materials to shipping out the door. So apart from the dozen or so who qualify for maintenance where is everyone else going to work?
Re: (Score:3)
While basically correct, you're looking at the past with a telephoto lens. The beginning of automation saw 70 odd years of chronic under-employment, that's 3 generations where sure, there was a bit of work putting up lights, but not enough to raise a family out of extreme poverty, even with every member out hustling gigs. Eventually the work did come back, helped by massive amounts of land being basically given away in the new world.
A hundred odd years ago when there was another spike in automation. society
Re:Yang will be disappointed (Score:5, Interesting)
The main problem here is that Boeing tried to introduce it for a product which was never designed for it - the 777 was designed to be built by hand in the 1990s, and you cannot change major structures to make it easier for automation without having to re-certify the aircraft design, which means big bucks.
Modern aircraft designs, such as the A380, A350 and 787 all have significantly automated manufacturing processes and designs to accomodate that, but the 777 and large parts of the 777X are stuck with the 1990s designs.
Re: (Score:2)
The A320 design is from the 1980s yet Airbus has managed to automate the assemble.
Re: (Score:2)
Not anywhere like to the extent that the A380 and A350 lines are - the A320 lines are still heavily manual labour orientated, in the same way that we are talking here. Airbuses gains in the A320 line is down to “pre-stuffing” major subassemblies built offsite and joined on the FAL - something Boeing proudly introduced with the 787 decades later.
But make no mistake, the A320 lines are still very very manual.
Re:Yang will be disappointed (Score:4, Interesting)
Your information seems to be out of date, Airbus has automated far more than just pre-stuffing. Joining and drilling is also automated, this is how they have managed to make 60 A32x per month two years earlier than originally projected.
Re: (Score:2)
I will have to disagree with you there - FAL joining has always been partially automated, and was intended to be automated from the very start in the 1980s. Airbuses increase to 60 a month is down to a lot of factors, the main one being that they have excess capacity built into their supply line to allow increases such as this (a rate increase to 60 a month has been mused internally since around 2010).
There is nowhere near the automation on the A320 family lines like there is on the A350 lines - the A380 i
Re: (Score:2)
They just need better robots. Not everyone will succeed in automating their processes. Those are the businesses that will be wiped out by those who succeed in full automation.
Re: (Score:3)
Automation doesn’t always lead to the creation of a post-labor society? Who knew.
But this situation is one of the *absence* of automation, so how can you infer anything from it if your antecedent is false?
What did they expect when they hired... (Score:5, Funny)
an alcoholic like Bender Bending RodrÃguez to work their line ?
Can't Automate After the Fact (Score:5, Insightful)
Tesla (Elon Musk) learned this the hard way. If you want to go with advanced automation you have to design a product with the limitations of robots in mind.
Maybe on the next-gen clean sheet product Boeing designs can be designed to be built by robots more easily.
design the product for automation (Score:2)
This.
You have to design the product for automation, not design the automation to make the product.
Re: design the product for automation (Score:2)
There's a fucking shocker.
Re: (Score:2)
Yes you can. This is what airbus did with their highly automated A320 assembly line. They had some teething problems last year but they have managed to overcome them.
Re:Can't Automate After the Fact (Score:5, Insightful)
I've noticed a lot of American manufacturers expect automation to replace all their labor, rather than change their cheap or "high cost" labor to highly skilled people to allow for automation to actually work and improve the process for automation and make a product FOR AUTOMATION (And improves our society overall as we get more educated people), which in fact is what drives your costs downward as your productivity rises. Instead, we have a driving force downward to Idiocracy with US companies that wants to get rid of humans completely, even the maintenance personnel (Competent people to fix machines? No way anyone is going to be making more than my sales guys on the factory floor!). The first thing I've noticed from that Boeing article is the talk about replacing all the people with hand tools with one robot without hiring someone competent to make sure its running right. Their whole mentality is about cutting people off and kicking them to the curb.
I've been to many plants that automate and of all the ones I see failing miserably are in the US. They end up having to hire operators and maintenance personnel from Europe to run the damn things because most managers expect Juan to be able to operate the whole thing for minimum wage without any supporting staff behind it.
Re: (Score:2)
Sounds about right. Automation is supposed to enhance productivity instead of replacing labour.
Re: (Score:2)
Exactly. Too often management forgets that - the automation is for the tasks that are ultra boring and ultra repeatable - take a block of material, carve out a slot in it, repeat on the next block. Or take a block of machined material and deburr it.
It's stuff the skilled workers hate doing because really, it's boring work and brings nothing to the job. But machines are poor judges of surface quality - if you need something machined s
Re: Can't Automate After the Fact (Score:2)
Re: (Score:3)
Well the "limitations of robots" is a moving target. I've been teaching myself quite a bit about GANs and machine learning lately and while it's not intelligence in the sense of thought it's extremely impressive how far we've come just in the last 5 years when it comes to recognizing varieties of the same thing, like faces in all kinds of poses and expressions and angles with different obscurations and lighting and facial hair and even across time. In many of these benchmarks we're now looking at beyond-hum
Re: (Score:2)
Re: (Score:2)
I imagine the scene more like this....
https://www.youtube.com/watch?... [youtube.com]
Re: Reached for comment... (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: All these robots (Score:2)
Re: (Score:2)
We're all doomed now (Score:1)
Well, I hope you like living out your life in a pod connected to a VR simulation of Earth, because this is how The Matrix started. Humans start taking the robots' jobs, and then the wars begin.
There is no silver bullet. (Score:2)
Full stop.
Silly them (Score:2)
Bots gonna take all our jobs (Score:2)
On the other hand, when you do get it working, the rewards are usually worth it.
Pilot testing ? (Score:2)
Re: (Score:2)
From the linked info in the main article, Boing seems to have developed the process at a former boat manufacturing in secret.
So they had a pilot plant. But the big question then is, was it still flawed at the pilot plant when being introduced into real production, and if yes, why would it be pushed into production?
And when a worker in 2016 mentioned that the process was horribly flawed, normally in automated manufacturing huge effort would be undertaken to understand the issue. And those robots must have se
the thing with robots. (Score:2)
Re: (Score:2)
when a robot makes a mistake it keeps making that mistake until a human corrects it. when a human make a mistake they can self correct.
There's a comic that disagrees with you.
https://xkcd.com/386/ [xkcd.com]
People don't "self-correct", they double-down on their mistakes because they don't like being wrong. It hurts their pride too much.
Fuck seattle times (Score:3)
Typical response from skilled labor (Score:5, Interesting)
I've been doing automation for 20 years now. Every skilled labor employee thinks their job is literally an art form, and won't believe you could automate it.
20 years ago, tool and die workers were telling us that a CNC machine could never do as good of a job as a skilled worker with a handmill. There's just a "feel", you see. You have to "listen to the cutter." That was after CNCs were already in wide use.
Of course what it took was some determined mechanical, electrical and computer engineers to prove them wrong. As an engineer, nobody bashes engineers more than skilled trades. It's inevitable. An engineer is constantly doing something they've never done before, constantly tries things, fails, iterates, and tries again. The apprentices of skilled trades, on the other hand, are immersed in a culture where a mistake means you're a failure. They relentlessly berate each other about the tiniest mistake. That makes sense when the success of your job depends on following strict methods and procedures that are time tested and proven to work. Unfortunately it also leads to stagnation in the trades. Nobody is comfortable trying something new for fear they'll be tied to a mistake and never live it down. Of course some automation projects are failures. They're hard, and nobody's ever tried this before. Good on the company for putting some money into R&D. But the guy with the pop riveter whose job might be replaced? He was likely hoping it would fail before it was even tested.
Re: Typical response from skilled labor (Score:1)
20 years ago, tool and die workers were telling us that a CNC machine could never do as good of a job as a skilled worker with a handmill.
And npw industry is desperate for tool and die people like never before.
Any more of your own [attempted] points you'd care to sabotage? I'm willing to let you do all the work while I sit back and chuckle.
Re: Typical response from skilled labor (Score:2)
Re: (Score:1)
What a missed oportunity (Score:3)
Watch out, Seattle! (Score:3)
With this blot on their downloadable resume files, the Boeing robots are not going to find jobs anywhere else. Wait until they start accumulating on city streets, guzzling cheap machine oil and begging tourists for AA batteries.
FUBAR (Score:3)
Boeing culture (Score:2)
You have to understand Boeing's culture to see why this most likely failed. It isn't one big well-oiled organization that works together to solve problems and innovate. It's a bunch of little empires run by the lower level managers who either do or don't get along. And if they don't, the users of the work product put out by one recalcitrant group will just have to be fixed. Because you aren't going to move people out of the way until they retire on their own.
Automation needs to be implemented from the grou
The assemblers are not machinists. (Score:2)
Machinists make things; assembly is done by assemblers, maybe millwrights.
For the /. crowd: It's like saying a programmer does networking. I'm sure some do, but it's not the same thing.