Slashdot Mirror
New 'Tent' Assembly Line Is 'Way Better' Than Conventional Factory, Says Tesla CEO (arstechnica.com)
A few days ago, Elon Musk announced a "new general assembly line" made with "minimal resources." As Ars Technica reports, this new tented facility "is seemingly the first phase of an entirely new building, dubbed 'Factory 2.0.'" From the report: The tent is easily visible from the nearby Warm Springs BART station platform. When Ars visited on Monday afternoon, there appeared to be cranes and forklifts moving around the site. We could not easily see inside the long white temporary structure, but there did not appear to be any newly completed vehicles rolling off the lines in the adjacent parking lot. Still, one automotive expert that Ars spoke with said that a new temporary manufacturing facility on the same site as conventional automotive factories was unprecedented in the industry. Dave Sullivan, an analyst with Auto Pacific, told Ars that he wondered what was wrong with Tesla's existing facilities, if Musk decided the company needed more capacity. "It's almost a sign of desperation," he said. "It's a sprint to be profitable in the third quarter." Ars notes that "each tent is 53-feet-high by 150-feet-long -- there seem to be several connected in a long line, mounted with aluminum framing." In a tweet, Musk said: "It's actually way better than the factory building. More comfortable & a great view of the mountains."
So, Elon Musk and his company just built in 5 weeks, literally from the ground up, what GM might spend 10 years architecting and designing, getting the EIR, hiring the right unions and negotiating their contracts, contracting for an acquiring equipment, actually assembling it, staffing it, and starting it running.
This race for profitability is actually viewed with contempt by a commentator of the conventional auto industry.
There is a lot to be said for agility, as any manager or investor in a start-up will tell you. This is simply another difference between Silicon Valley business and the conventional auto business. The same sort of difference that allowed Tesla to make electric cars that could actually compete with gasoline cars on their own turf.
I hear Ferrari is just starting to introduce one that might compete, after Tesla has had theirs on the market since 2007, with three new models since. But that's it so far. Ferrari is pretty agile as auto manufacturers go, but at the cost of low manufacturing volume. Ferrari won't be selling its cars at even 5000/month. Tesla is trying to get both the volume and the innovation.
Bruce Perens.
I think the real problem is they did it too quickly. Too much means that some things can never be automated. Too fast just means they have to take longer to get it right. Some lines that were made manual are now automated again, as they got the process working.
The main point of manual labor is that you can speed up by hiring more people. With automation, you have to make the machine actually work before you can scale up by buying more machines.
Bruce Perens.
I've never been in a modern auto factory, but I've spent a lot of time in places where complex weapons systems were assembled. Temperature and humidity control wasn't an issue. In California, there wasn't any of either although the sheer mass of the facility tended to moderate temperature swings. In colder areas elsewhere there was some heating.
We're not talking a semiconductor fab facility here.
You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
Tent in tent, repeat as required. Been in several aerospace composites factories that are little more than tents-- a Butler Building really isn't all that different. Usually the floor slab does a lot to moderate temperatures, but you can easily set up a modular clean room inside if you need to.
Have you guys ever been in Denver International Airport main terminal? It's a tent. It gets as hot in Denver as it does in Fremont.
Snark all you want but just because something is a quote TENT unquote doesn't mean it isn't a robust and practical structure.
The paint shop would most likely be in the existing structure, not the tents. But there's really no reason why you couldn't put it in a tent; all you need is a positive pressure system. Mobile chemical warfare decontamination units use tents; you just plug some air pumps with filters into them and make sure the pressure inside is higher than the pressure outside. Air will keep leaking out, but that stops any contamination from leaking in.
Since most of the robotics come from Germany, I am guessing that the next factory will be in Europe, possibly Germany or France. It should be easy to set up a quick and inexpensive tent, followed by robotic lines.
I prefer the "u" in honour as it seems to be missing these days.
Congratulations, you're the thousandth immature male to make that joke :P
Concerning the article, some of this is head-smackingly stupid. E.g.:
*facepalm*
There Is No More Space At Fremont. Something that has been discussed endlessly in the conference calls and at the investor meeting. Tesla applied for a permit long ago to build a new building, but it's still in progress, and meanwhile, they've been filling up Fremont at a rapid clip as they expand Model 3 production alongside the existing S and X lines. They could build a new line in a temporary building, get it up and debugged and running and turning out vehicles, or they could sit around waiting for months (or more) for a new building. Gee, I can't imagine what's the right choice here, hmm...
The current building (still being improved, but with the line in place inside) is permitted for six months, but given that they've been liking it, it may become permanent.
I was watching this thing on TV about some guy named Hitler. Someone should stop him!
Um, everyone who listens to the conference calls and the investor meeting knows? Just because you don't know something doesn't mean that nobody does :)
You want an example? Flufferbot. There's a loose fluff used in the battery packs, and they made a robot to place it. Now, if you're trying to come up with something that would be difficult for computer vision systems to process, and for robotic arms to handle, you couldn't do much better than "fluff". The robot spent most of its time finding new and creative ways to fail to pick up the fluff, as well as to put it in inventive new locations. So you had these expensive robotics technicians both bailing it out of its mistakes and trying to adjust its programming to prevent them, on and on for months on end. When you could just have simply paid people to place the fluff.
Flufferbot is gone, by the way ;)
I was watching this thing on TV about some guy named Hitler. Someone should stop him!
Meanwhile, if you actually hung out on Tesla forums and watched people take delivery and write about their cars, you'd know that this is a myth. Yes, as with any brand, the occasional vehicle has problems, and needs to be corrected by a service centre. But it's not at all like the shorts portray, in their endless glee sharing every last case that they can get their hands on.
Tesla has always had the highest consumer satisfaction rate in the auto industry. Back in the Roadster days when they were selling hundreds of vehicles, you all said, "Well, that's only because you have hardcore early adopters - once you get more mainstream, people will stop putting up with it and will hate Tesla!". Then Tesla started selling thousands of vehicles with the early Model Ss. And you all said the same thing. Then they were selling tens of thousands of vehicles. Then a hundred thousand per year. And are now moving into the hundreds of thousands per year. At what point is your "people are going to start hating Tesla" hypothesis going to come true? Did you ever stop and think that the reason people tend to rate them well is that they actually really like the cars after having owned them?
I was watching this thing on TV about some guy named Hitler. Someone should stop him!