Interesting analysis. Michigan has the lead.
Michigan does have a lot of robots. I suspect that the numbers in the article are actually very under reported and don’t take into account older robots at smaller manufactures. Watching the hundreds of robots move in a modern automotive bodyshop is truly amazing to see. Anyone coming to Detroit Champs next year should check out the Rouge Factory Tour.
http://i.imgur.com/OClv2Ua.jpg
Interesting article. Thanks for sharing.
Just to set expectations:
The Rouge Factory Tour is awesome, but unfortunately you won’t see the body shop. You will see automated moonroof install though.
I was lucky enough to see the Dearborn Truck body shop during training to work the line in final assembly for Ford’s college graduate new hire rotation program. Almost everything there is automated, including moving parts and materials around with automated guided vehicles. Here is a cool time-lapse from when the F-150 changed to an aluminum body. The time from the last steel truck made to first aluminum truck made was around [strike]three[/strike] 11 weeks.
Looks like Indiana is up there too. I play with such robots (Motoman, Kawasaki, and sometimes Fanuc and Panasonic) for a living, so I shouldn’t be that surprised. Nothing like the massive assembly lines with hundreds of robots, but a robot is a robot in this case.
I can imagine what a mess the F150 model change would have been. Aluminum, in sheet metal/body shell application, is a different animal. It doesn’t stamp the same or weld the same. Heck, aluminized steel (aluminum coated sheet steel used in exhaust heat shields) is bad enough to spot weld. Transitioning to aluminum (not counting the entirely new body designs which would mean new tooling and robot programs at a bare minimum) in 3 weeks would have been a major effort, although I’m sure they trialed the new setup in some capacity beforehand. Toyota does so for about a year before official product launch (and as a supplier, my employer does the same).
Well I guess west coast isn’t best coast… 
For once 
Turns out I was off on the timing, it took 11 weeks. The aluminum body is almost all riveted and glued together. There was absolutely a ton of R&D and trials before mass production started.
Not sure what you mean, looks like some of the largest and darkest circles are in west Michigan. A lot of great work going on there, it just often goes unnoticed since it’s not all automotive.
He’s just repeating an often-used comment by people from a certain western state… mostly used by folks who have never gotten outside their bubble and have very little life experience 
I could step outside my office and take a picture that looks very similar to the Rouge rebuild. Just finished the concrete pour and they are cleaning and redoing the roof and replacing the lights before wheeling in all the CNC’s. Start to finish will take about a year. The robots and automation used are amazing. I wonder if they are including multi-axis gantries as robots.
I think you missed the joke. I’ve actually been to the West Coast of Michigan (on one occasion). Why yes I’m from CA, how did you know?
I fault this study because it only focuses on the classic industrial robot. I believe they should have looked at the larger picture and studied production automation. The whole chemical industry is left out of this. Add in agriculture and food production automation and that map would change significantly. This would allow a more realistic nation wide economic study. The automation revolution in this country is well established and the economic effects largely already run it’s course. We need to start looking to the future and study how the AI and BI technology will impact the future of our society. We are about to face a major upheaval. Automation killed the blue collar factory worker, AI will decimate the well paid mid level white collar worker.
I was out to dinner with my son when his personal and work phone went crazy. It was one of his production lines letting him Know that a bearing in one of the grinder stations was developing a little to much vibration. Sent a predicted failure estimate too. Yes our machines are becoming intelligent.
I don’t know how many NDA’s were violated but, one of my customers let me have a conversation with Mr. Watson. I was a little disturbed by that interaction. I had no idea AI was so far along. AI is going to disrupt our society. How can we deal with it? That’s what needs studying.
Ah. Explains a lot. The same way aircraft and older aluminum bicycle frames/forks are made.
Usually at work each project has 5 dedicated rounds of trials (that relate to customer orders), with extra rounds in between for line improvements (cutting cycle time, etc.). We don’t get 11 weeks to transition though… following the last week of the old is the first official week of the new (with internal ramp up starting not long before to build inventory). Was this one a separate new line or a complete retrofit of the existing line?
This was a complete tear out and replacement of the existing line. Minutes after the last shift of the 2014 model, workers started removing the old equipment. F-150 production was shut down during the change over. Most new models have a smooth transition similar to what you said. Paint and final assembly did not require a complete retrofit.
I am confused by the conclusion of the article. I would think that in areas with high robot density, people would see they are not replacing but rather enhancing workers and jobs. Nobody ever lost a job they enjoyed doing to a robot - in fact the robot likely allowed the worker to advance to a more fulfilling role.
Experience and familiarity often ease fear, not amplify.
Those unemployed because of workplace automation may beg to differ. Automation makes manufacturing processes way more efficient. You can do the same job with fewer people. Those people that remain may, as you say, “advance to a more fulfilling role”… but what about those that get shown the door?
Take, for example, a welding job for an automotive manufacturer. You do the same thing all day, every day - weld two pieces together, and the two pieces are always the same, always get welded in the same place. Now they bring in a robot that can weld those two pieces together in half the time it takes you, and has fewer parts fail quality control. Does that worker magically go from being a welding expert to being a computer/robotics expert capable of running and repairing a dozen welding robots?
The introduction of industrial robots has put a lot of people out of a job. It’s replacing a lot of traditional job roles, and those with expertise in those roles find themselves with resume’s that can’t get them other jobs, now that their career has been replaced by robots. On the flip side, there are a lot of high-tech jobs out there that companies can’t find qualified people for. It’s a huge gap in education and training between the jobs that are being replaced by robots and the jobs that robots create.
In the long run, this gap will take care of itself. Fewer students will learn how to weld, and more will learn the skills needed for the newer jobs. But in the short run, you have an aging population with skills that are no longer needed and a difficult path towards gaining the skills they need to be employed, in a time frame that doesn’t leave them financially ruined and at retirement age. This is a real problem for many people, created by industrial automation, outsourcing manufacturing overseas, and a changing economy that is shutting down industries (like, for example, a decrease in coal mining in favor of cleaner, renewable energy).
Those areas with high numbers of industrial robots have gone through the pain and know it well, and I think it’s natural for them to wonder what jobs are going to be replaced next. Those areas without a lot of industrial robots probably just aren’t thinking about it - out of sight, out of mind.
I’ve toured dozens of manufacturing facilities and spoken with many additional company representatives. I have yet to have any conversations with any of these folks in which they did not underscore that no people were ever fired by the addition of robots. In some cases, jobs were lost due to attrition (employee retired, moved, etc.) but - and they are all crystal clear on this - nobody was ever fired.
I’m glad those companies could manage that… however research has shown that’s seldom the case. Here’s an article from earlier this year about a study from the National Bureau of Economic Research. That study estimates that 3 jobs have been lost for every industrial robot introduced. It says hundreds of thousands of jobs have been lost do to industrial automation.
I would like to counter your example involving welding robots. Wolf Robotics, who manufactures welding robots is based in Fort Collins and I have toured their facility a couple times.
From what I have heard, every time Wolf outfits a company with industrial welding robots the humans who did the welding keep their job. The people gain some robot management skills but really they don’t have to become robotics experts. Their job is to watch the robots, or in the case of welding robots, listen to them.
Welding experts learn how to tell if a weld is good by the sound it makes. A robot can monitor temperature, speed and a whole bunch of other things, but it can’t listen to the weld and tell if it’s actually good, the human who used to do the welding does that for the robot.
Obviously this relationship doesn’t occur with every industrial robot. The impact of this relationship is what I would really like to see a study on. How often does automation actually kill jobs and how often do those jobs just shift in subtle ways we don’t always notice (human goes from welding to listening, jobs move to places like Wolf, etc)
Edit: Jon’s article is just what I was looking for.
This has been closer to my (admittedly anecdotal) experience. In name, jobs were “lost” which would support the findings of the study Jon offered, but the actual humans who were performing these jobs maintained their employment by taking positions with different job titles, which would be in line with the passage at the end of the aforementioned article.
To what extent does this happen though? I’d imagine that you wouldn’t need as many robot-watching/listening related jobs as you’d originally have welders. I’m not sure if it’d be economical for companies to keep the same number of people employed, as they all need to be paid.