7 ms·

I'm skeptical about the timelines of these reports. Certainly we'll never see all 1.8 million drivers lose their jobs overnight. These guys are predicting auto-

by ufmace 10y ago

I'm skeptical about the timelines of these reports. Certainly we'll never see all 1.8 million drivers lose their jobs overnight. These guys are predicting auto-truck apocalypse in 5-10 years, and we still don't have a single commercial system on the road that's capable of even lightening the load on a driver, much less replacing him. I think truck automation will go in 2 ways at the same time, and we can watch the progress of each: Systems to ease the strain on independent drivers. Ones that can cruise on the highway without supervision indefinitely, but need help with city streets, parking, maintenance, loading and unloading, keeping manifests, dealing with whatever company is loading and unloading the cargo, etc. They may need somewhat fewer of this class of driver, since the trucks will be able to run more continuously and there will be less need for second drivers and probably fewer trucks. Motels and truck stops will hurt some when the truckers can sleep while the truck drives instead of stopping. I think we're at least a decade away from this existing at all, much less being common. Full automation for tightly integrated logistics chains. Maybe the Wal-Marts, Amazons, Fedexes, and other huge companies that own the entire logistics chain will be able to figure out how to use fully automated trucks, that can drive from one company facility to another, complete with parking and maneuvering, driving local streets, and letting other company systems handle the logistics of loading and unloading and keeping track of what items are where. I bet at least one of them will start experimenting with something like this in the next 5-10 years, but probably at least 20 years before it works well enough for them to cut down on the number of drivers they employ. There will be job losses, but it will be slow and gradual. There should hopefully be plenty of time for the economy to adapt, and hopefully either create new jobs for all of these people to do, or move towards something like UBI. I think we'll have to have a massive cultural shift before anything like UBI would be considered or even possibly make sense.

galdosdi 10y ago
I dunno. Even if "all" the first generation can do is freeway driving (and let's say, not even urban freeway, just "easy" rural portions of freeway) that seems like it still would make the vast majority of long-haul trucking jobs vanish -- just by definition of "long-haul." If you ship something from New York to Chicago by truck and it takes 14 hours, maybe the first and last hour or so are getting in an out of the source/destination cities. That leaves 12 out of 14 hours where the truck can drive unattended, eliminating the need for about 12/14 of the truckers (or their work hours) that drive that route. It depends on what proportion of truck drivers today mostly do long-haul as opposed to short haul. I have no idea.
- ufmace 10y ago
  I'm not deep into the freight industry, but I have a few acquaintances who are, and reportedly a lot of the actual shipments that take place are between unrelated companies who only moderately trust each other. The freight company is contracted by one or the other for the route, and usually the truck itself is owned independently by the driver, with the freight company providing the routes and organization of loads. The truck owner isn't going to trust some random other driver halfway across the country to drive his truck, even if he trusts the auto-driver to navigate the freeway safely. The freight company isn't going to just trust the shipper and receiver to load and unload the right stuff properly and not disturb anything else on the truck. The shipper and receiver aren't going to just trust each other and the freight company to ship the right stuff to the right place - they all want somebody in the truck there who knows what's going on. Basically, the drivers don't just drive, they're the glue that holds together a whole complex system. Even if we had a perfect auto-drive truck today, it would probably take decades to figure out systems that work well enough that shipments work right without somebody who knows what's going on in the truck. Like I said, I think the best we can hope for short-term is to take some of the load of actually driving off the drivers, leading to longer trips and fewer stops. That's also why I said that the only companies in a position to really use auto-trucks are those who already have integrated logistics chains, where drivers who work for the company drive company-owned trucks between company warehouses to move goods that the company either already owns or has taken responsibility for tracking.

by ufmace 10y ago

galdosdi 10y ago

I dunno. Even if "all" the first generation can do is freeway driving (and let's say, not even urban freeway, just "easy" rural portions of freeway) that seems like it still would make the vast majority of long-haul trucking jobs vanish -- just by definition of "long-haul." If you ship something from New York to Chicago by truck and it takes 14 hours, maybe the first and last hour or so are getting in an out of the source/destination cities. That leaves 12 out of 14 hours where the truck can drive unattended, eliminating the need for about 12/14 of the truckers (or their work hours) that drive that route. It depends on what proportion of truck drivers today mostly do long-haul as opposed to short haul. I have no idea.

ufmace 10y ago
I'm not deep into the freight industry, but I have a few acquaintances who are, and reportedly a lot of the actual shipments that take place are between unrelated companies who only moderately trust each other. The freight company is contracted by one or the other for the route, and usually the truck itself is owned independently by the driver, with the freight company providing the routes and organization of loads. The truck owner isn't going to trust some random other driver halfway across the country to drive his truck, even if he trusts the auto-driver to navigate the freeway safely. The freight company isn't going to just trust the shipper and receiver to load and unload the right stuff properly and not disturb anything else on the truck. The shipper and receiver aren't going to just trust each other and the freight company to ship the right stuff to the right place - they all want somebody in the truck there who knows what's going on. Basically, the drivers don't just drive, they're the glue that holds together a whole complex system. Even if we had a perfect auto-drive truck today, it would probably take decades to figure out systems that work well enough that shipments work right without somebody who knows what's going on in the truck. Like I said, I think the best we can hope for short-term is to take some of the load of actually driving off the drivers, leading to longer trips and fewer stops. That's also why I said that the only companies in a position to really use auto-trucks are those who already have integrated logistics chains, where drivers who work for the company drive company-owned trucks between company warehouses to move goods that the company either already owns or has taken responsibility for tracking.