I wonder if this is gonna be like Waymo, where statistically speaking, the amount of cars that they had on the road, with how many crashes they had, it was deemed ineffective and dangerous, but as soon as they reduced the cars on the road they had less incidents because less odds. At this point, they probably continue to suck up subsidies and donar $ so they keep their little goofy business afloat. So what happens now? When It crashes into a school bus, who is held accountable? This is not a good idea. We need to tax the rich and corporations shouldn’t have so much power. We didn’t ask for this future. Anybody with a damn lick of sense knows that this is a stupid idea. Also, why? Like, just make more trains. America is so stupid. The fact that you just fight over cars versus walkable cities. I am actively trying to find a way out of here. These are horrific insanely stupid ideas. It’s like doing it the hard way because you’re too prideful to admit that you have a shitty infrastructure.I have seen Europe, I have experienced it, and it is far superior to the shitty infrastructure of the United States. I was born in America. I lived in America, but I do not identify with this way of life or culture. My head spins with just where people’s minds are at in this country. How little they know. It’s terrifying. If you’re out there, just know. It is better elsewhere and chase those goals. You’re not crazy. There’s a better world. Not perfect, but a better world out there. Leave if you want to leave, don’t let anyone tell you otherwise.
Good. No more freedom convoy MAGA base.
Both political parties are anti-union, anti-worker, and anti-American.
Not after the first snowfall, they won’t.
Great, just in time for the number of shipments of imports needing to be distributed across the US to plummet…
Americans will do anything to avoid just using trains.
Trains help poor people too. We like to pretend we don’t have poor people. Makes them easier to ignore while pretending to be Christian.
Except that nearly all US rail is for freight. We hate PASSENGER trains. We freaking love freight rail.
Except that’s rail only carries 16% of freight by weight and 2% of freight by value.
Pretty sure USA hates freight rail too.
You’re looking at a different issue. I’m referring to passenger trains vs freight trains and you’re talking about freight trains vs semi trucks. I’m saying that the rail we do have, we overwhelmingly use for freight. It’s the primary reason we still have trains today in the US.
In regards to percentage of freight shipped by rail vs other means, I believe you that semis take a ton of that.
But american freight trains are laughably bad too
Yes, but “we will avoid trains no matter what” is blatantly false. It’s terrible, but it is our main method of shipping freight from ports to inland cities.
And semi rigs (which are the topic of this post) are…personal transport?
No, the subject is shipping cargo. Try to keep up.
You were the one that brought passengers into the conversation.
Yeah… Saying we don’t use them as much as… Freight. Try to keep up?
While I don’t necessarily disagree with you, trains are used here all the time specifically for long haul stuff.
I used to be the shipping/receiving guy in a warehouse, it fell to me to arrange all of our freight pickups, which was annoying because I didn’t really have direct access to any information about pricing, deadlines, etc. so I was constantly going back to the office to show someone quotes to see whether the rates and transit times were acceptable.
Most of our freight was LTL stuff (less than truckload, a couple pallets, not enough to fill a truck by itself) but a few times every month or two we’d get full truckload sized orders.
When it came to them, often “intermodal” shipping had much better rates. Intermodal meaning at least 2 different forms of transportation were going to be used. Truck, train, boat, cargo plane, etc.
As a US-based company with mostly US-based customers, that usually meant rail for us.
However, almost none of our shipments went intermodal because it was too slow for our customers.
It wasn’t usually a drastic difference, we’re talking maybe 1-3 extra days in most cases. Over the Road (OTR) there weren’t many places in the US that we couldn’t get freight to from our location in 5 days or less, and those 5 day locations were mostly real middle-of-nowhere customers on the other side of the country.
It always blew my mind that we didn’t or couldn’t push our customers to just place orders 2 or 3 days earlier to save some pretty significant money on shipping.
I don’t claim to know much about the industry, i was just some kid who needed a job and ended up the shipping guy because I knew how to use a computer and spoke English. But we a textile company that made things like work clothes (chef coats, scrubs, industrial work wear, etc) and restaurant table linens, and we sold mostly to bigger wholesalers, business service companies, etc. who would resell it or provide it to their customers as part some sort of contracted laundry service or something, so not really something I’d think of as being particularly time-sensitive or wildly unpredictable that they couldn’t anticipate their bigger orders a couple days ahead of time
Guess it probably says something about how much we all love instant gratification.
Inventory became evil decades ago. “Just In Time” logistics became the norm instead of having warehoused inventory on hand. The beancounters all decided inventory was money that was sitting around not doing anything and maintaining the warehouse space cost more too. Can’t have those costs on the balance sheet. So speed in receiving smaller shipments more often is now the norm, along with ordering when you need them instead of ordering ahead of time, because some beancounter isn’t gonna be happy about extra inventory.
The beancounters are right about the costs. What they’re not right about is the risks. JIT supply chains are much more fragile, and to achieve some degree of resiliency, even sophisticated manufacturers will often mantain stockpiles of some critical goods. And things get even more funky when there’s only one good supplier for something, or the cost of switching suppliers is high.
as these tariffs start kicking in, companies are really going to regret not having local inventory.
Worked in two factories since Covid. The first stockpiled components we produced in house, and relied in JIT logistics for external components. Which was basically the stupidest arrangement they could have cone up with. They had 10+ years worth of parts they could make in house, clogging up their warehouse. And couldn’t ship anything because they were waiting on suppliers.
The other built two new warehouses to stockpile external supplies, and never let up on production.
Yeah, if you ever need stories on just how stupid senior managers can be, look at supply-chain case studies. And don’t blame the accountants: it’s their job to report costs, but it’s the job of the managers to deal with risk. And running ultra-lean JIT comes with the risk that a five-minute delay in some critical component can shut down your line. It’s not the beancounters’ job to have appropriate plans in place to prevent that from happening. It’s the biz-school bell-ends who are asleep at the wheel or thinking that they’ll just pretend there’s no risk and hope they’re lucky enough to translate those low running costs into their quarterly bonuses. And the contingency planning if the supply chain does glitch? Often it goes no deeper than having a scapegoat lined up.
It’s always the fucking accountants.
Rail is used in the US. We just don’t have as much rail infustructure so they can only get so far. If the port/factory/wearhouse aren’t connect by rail then they’ll have to use trucks for at least part of the transit.
Probably could have built a lot of rail for the cost of R&D on self-driving semis…
I’m not so sure. Infrastructure is hella expensive and the US government already maintains the highways that make trucking make sense.
Not necessarily. A 40 tonne lorry damages the motorway as much as 1000 passenger cars. It will lead to the state having to renew the road surfaces every few years. Rails don’t have that problem, they’ll happily take 100 tonnes for decades.
The point I’m making is that the government has already decided to maintain the highways, so continuing on is the status quo. If they wanted to make new railroads they’d have to expend political capital to get anything new funded.
Maybe 2 or 3 single rail lines across the country.
You guys gotta remember that the US is double the size of the entire EU. I will say that I don’t disagree in that more rail would be nice, but you have to think about this logically.
Oh I do, it’s where I live. At current costs its about $1.6m(1) per mile, so yea, agreed, probably not much. Will have to check back in 5 years after we see the costs to operate and lawsuits from accidents 😆
how about historically? we had rail, and it was great. Most of it was ripped up at the behest of auto manufacturers.
It was mostly the street car systems that got ripped up, not the stuff that carries freight.
trams were the biggest casualty, but not the only.
Trains are great but they don’t typically run to your local warehouse…
They used to.
Because the warehouse was built on the tracks. Alas that infrastructure tie-in has mostly gone away, new facilities are built with proximity to cheap labor, land, and easy to consume + pollute natural resources.
They have, and they could again
Canals can be useful for this as well, Lowell MA used to have a huge industry all on waterfronts
Could but don’t.
Not much competition in railways. Like literally none.
Why not make automated trains with their own dedicated right of way?
But that would require investment in infrastructure…
Bet that semi trucks are more expensive due to road damage and congestion alone.
Yes, but that’s all subsidized by taxpayers, so it’s more expensive overall but cheaper for YOU.
Privatize gains, socialize losses. The CapitalistTM way!
Efficiency, pollution too (even when electric, because tires and break dust are a thing)
What you don’t get is that trucks last less and consume more, therefore it’s better for the robber barons.
*** everyone but the lobbyists liked that ***
They already are automated trains on freight only routes like mines.
Outside of mines or just in mines? I know that mines are becoming more automated but what about commercial routes.
It’s absurd to suggest running a railway to every warehouse in East Bumfuck, Missouri.
No one’s claiming that. Trucks can still handle the last mile just like they do it with container ships.
Im no logistics expert byt ship -> train -> semi sounds like a great infrastructure design especially now as the container is interchangeable between all of these mediums.
Oh. But a road is famously cheap.
The road already exists
Just one more road and the traffic will get better /s
Compared to building and maintaining a railway, yes, by orders of magnitude.
A road built and maintained by taxpayers is much cheaper (to a shipping company) than building, maintaining, and operating a railway. Making taxpayers responsible for the infrastructure you use is one way to make your business much more profitable.
Citation needed
A cursory search shows rail in rural areas is $2 million per mile and a highway is $4-10 million per mile.
Yeah but it’d be fucking insane to build a state highway to each and every destination in every hamlet, just like it would be for rail.
And it’s not just cost of initial construction, it’s also cost of maintenance. If the ground shifts slightly under the road, it’s a bump. If it shifts under a railway, it’s a derailment for the first train that finds it and a couple million dollars in recovery and repair, plus the downtime while that section is out of service. And that doesn’t even start to account for overhead like signal operation, whereas on a road you just use a stop sign.
I like trains more than the next guy, but you absolutely cannot just replace every road with a railway.
I think you’re missing the general point.
In the cases you’ve described, having automated semis would not be feasible. Automated cars already have a hard time in San Fran and AZ cities with smooth asphalt as it is.
The places where automated semis make the most sense, i.e. large, well maintained highways connecting large urban centres, can be better served with automated railways.
The engineering is much simpler, fewer degrees of freedom and a much more constrained problem space (and hence constrained solution space), for automated railways than highways. Creating a safer environment for all. Also not having to deal with semis as an individual driver.
Railways (funded through private investment): https://www.aar.org/wp-content/uploads/2025/01/AAR-Rail-Network-Map-2025-1.jpg.webp
Highways (publicly owned, operated, maintained): https://www.fhwa.dot.gov/planning/images/nhs.pdf
There is some good coverage with railroads, but as you said not nearly extensive as the public road network. But I bet you the vast majority (above 60%) are along corridors with railways. However two big hurdles need to be overcome, greater investment in throughput capacity and the fact that trucks can go from ware house to ware house.
However both issues can be solved.
The places where automated semis make the most sense, i.e. large, well maintained highways connecting large urban centres, can be better served with automated railways.
On this I agree. For popular, well-defined routes, rail absolutely makes sense, not just for freight, but for passenger transport as well.
As of Thursday, the company’s self-driving tech has completed over 1,200 miles without a human in the truck.
That’s not an impressive number. That’s like 2 days’ worth of driving.
Yeah that’s about 2 and a half round-trips between Dallas and Houston, that’s…not a lot to be calling this thing ready to go and pulling out the safety drivers.
I wonder how these handle accidents, traffic stops, bad lane markings from road construction, mechanical failure, bad weather (heavy rain making it difficult/impossible to see lane markings), etc.
You’d think they would be keeping the safety drivers in place for at least 6+ months of regular long-haul drives and upwards of 100k miles to cover all bases.
The one article I heard on TechLinked talked about them using lidarr.
So better in every way than a tesla.
Assuming they are top mounted, they have a better scanning coverage than a regular car.It would be more interesting to know how many miles they completed with the safety driver in the vehicle.
That figure is without a human in the truck, not with a safety driver. I.E, they’ve done a bunch of testing beforehand.
Most rigs go at least 1,000,000 miles and that isn’t isn’t even end off life. You’ll be paying not much less than new for a rig that only has 100k, that’s practically brand new. These systems should have 100 million proven miles. These things weight 80,000lbs which can be very hazardous materials.
You should see the pile ups semis cause in low visibility. Even with really good lidar, I hesitant to say autonomous trucks can be safe running off independent systems on full mixed use roads.
We could add those systems to all roads to feed back to semis to know conditions and hazards miles before they reach them. We could build new smart roads for all autonomous vechilce to travel on separately.
Or we could just end the 100+ year old railroad cartel. Could move people and cargo with ease. But that isn’t profitable.
Its enough to prove the concept.
If it saves 1% of operating costs trucking businesses will be falling over themselves to implement it.
Nah no one’s going to bother with new tech for 1% - that’s crazy.
Great… I can’t wait to be hit by one of those on my motorcycle
I’d actually bet they’re safer than some tweaked out dude on his 20th hour at the wheel.
Ok, elmo.
Same. Our government can’t even figure out a way for us to trigger a green light so I’m not confident that any self-driving vehicle regulations will consider us either.
Heh, I got hit by that stupid thing today. Luckily the crosswalk button was right there, so I ran over and smacked it before the traffic signal cycled again.
I’ve heard of people doing this but my strategy is to just wait until it’s safe and run the red light or go right and bang a uey.
I couldn’t do that on my 49cc scooter 😅.
I’ve got a 1980 Honda XR500 as well, but it needs some work (and tires, badly) before it’s roadworthy again.
Large neodymium magnet on the bottom will do it. Most are induction activated. They taught this in every motorcycle driving class I ever attended, along with the rules for legally running a red light.
No, it won’t. I invite you to get some large magnets and place them directly on the loop cuts in the street.
Well yeah, they have to move over the sensor loop. You can’t just place it in the center. It’s part of a process that happens as, you know, you pull up the the light. They are a little fancy.
That way you’ll also automatically collect all sorts of valuable metal treasures along the route. For free!!
At least the screws won’t collect in your tires! Lol
I vaguely remember a dystopian book that described that exact thing as the protagonist thinking he was looking at an odd flag on the front of the truck until he realized what it was. Can’t remember what the book was though 😔
What an incredibly infuriating waste of effort that would be so much better spent on trains, driverless or otherwise.
I disagree. There are many situations where a truck is better suited for transport than a train. The US already has a pretty large freight train network. I agree that there definitely should be more investment in rail as well, but there’s no reason for both not to exist at the same time.
There are, but “long haul routes” are definitely better for a train.
I don’t know why you’re being down voted; here’s an upvote for being sensible.
Fuck cars. Fuck massive death trucks even more.
And how do they handle a person slowing down in front of them and hijacking them? At least a human might be able to navigate away aggressively but I think the programming would prevent as much harm as possible.
This new lawless future and we may need to raid corpo lords.
I can’t really imagine people wanting to hijack a truck that’s basically a giant camera and tracking system.
I dunno I can see it being done, go in first with a drone and blind the cameras one by one (not hard to rig it up with spray paint) then grab the goods, pick an isolated section of the route so you’re gone by the time anyone comes looking.
I for one predict a glorious era of road pirates.
I’ve seen plenty of youtube videos to know people are dumb enough to try this.
The resistance needs supplies!
Really?
Big box truck, no plate or other #s on the rear. Halloween masks for the crew.
I think that’d work, no problem.
If they unloaded the goods to another truck really fucking fast, maybe. Otherwise they’re just being recorded on top of a giant GPS beacon that’s alerting a system that something abnormal is happening so rally nearby law enforcement. It’d way easier to hold up a bookies, with a shotgun, Saul.
I think the programming would prevent as much harm as possible.
well, yeah… why wouldn’t you want a human to do the same thing??? you’re watching too many fast and furious movies.
Firstly, no one in an 18 wheeler loaded with cargo is “navigating away” from anyone desperate enough to attempt such a scheme. This entire idea is ludicrous, think about how slow and massive those trucks are.
Secondly, you don’t want an 18 wheeler loaded with cargo being driven aggressively. You’re just escalating the risk of killing yourself and everyone around you, for what, a truckload of insured corporate assets?
Honestly, sounds like the corporation’s problem. I’m more afraid for human lives than some product in the back. In a case like that it’d be better to not have a driver who could be killed.
Why do you think a morbidly obese truck driver would have any luck against blaggers with shotguns ?
Driverless does not meant unmonitored. Aside from numerous sensors, including door sensors, you really think if it suddenly slows to 0 mph at an unscheduled time/location that it’s not going to alert someone? “Hey, your freight just stopped transporting itself. Guess we should do nothing”. Aside from most of these being ready to be taken over by a remote driver if need be for liability and convenience reasons.
But, do they speak English?
Even in a hypothetical best-case scenario world, unless you have a driver on board any malfunction and you’re delayed 2-8 hours because there wasn’t a person in there to repair anything
How many issues on a truck is the driver able to fix themselves though?
A lot. Most of them do a lot of basic maintenance and break-fix work themselves.
Lol, no they dont. This is such reddit shit. Say something people will believe for the updoots because they have zero clue.
Your average driver is lucky to have a bat to check tire pressure. All the old guys that know how to work on their rigs are now too old to do it, or have enough money to just call the roadside desiel mechanic. 90% of drivers don’t own their rig, don’t give a shit, and are taught to just call the company to send a mechanic.
Yeah, I’m a driver, and I replace lights and fuses, fill up tires and fluids, change wipers, and that’s it.
Anything else wrong, it’s mechanic time.
Nailed it. If you don’t, you’re stuck waiting on service to replace those menial things for you, putting a 2 to 8hr dent in your workday, delaying all of your schedules. Source: operate a small trucking company.
Terrifying.
I wonder how much our car insurance will go up due to this.
I get that you’re scared about multi ton vehicles running without a human. But self driving can and actually be safer than human drivers sometimes. Yes, self driving vehicles can cause devistating accidents in situations where a human driver would have handled the situation much better. Sometimes they can just bug out, which seems particularly dangerous, but we also need to consider who they’re replacing: Humans. Humans get tired, Humans text & drive, Humans blink, Humans Yawn, Humans do drugs, Humans sometimes just don’t pay attention. Because machines don’t have any of these factors, they can statistically be much safer, of course assuming the technology is ripe enough and thoroughly tested before it’s used.
Ok, elmo.
