Memes

US freight rail looks great since for one, the freight railways dominate the scene, and for two, the US is up there in network distance as well as cargo transport volume in tonne kilometers. And of course, they have some very high operating margins.

However, the devil's in the details.

For one, if we redefine the amount of cargo transported to be measured in US Dollar kilometers, they're suddenly doing a trash job. Much of the cargo they move is fungible (it doesn't matter what unit of this cargo you have, any kg is a good kg), bulky and not time sensitive. Things like coal, crude oil or gravel are disproportionately common freights on US rails, compared to other places.

Secondly, they put a lot of trains besides the tracks. I recall seeing they managed to derail about 1700 trains a year. Most other train systems don't even do a tenth of that in a decade, even when corrected for track mileage.

Speaking of track mileage, US railroads actively reduce the amount and quality of track, while bitching & moaning to the government and the press that they're overburdened. Meanwhile, they also operate a procedure of precision scheduled railroading, which I'll spare you the details on, but let's just say it's not precise, it's not scheduled and it's barely railroading, and despite forcing some train crews to sit back and do nothing for hours, it still saves them one train crew. The only time they'll actually expand is because either they really did have a bottleneck for decades, or something catastrophically fails.

On top of that, the freight railroads do everything in their power to avoid capital spending, so they refuse to electrify their lines and/or to install more advanced signalling and train protection. One major fuel shock, and American railroads are on their knees while India, China and most of the EU are laughing. And most signalling is unenforced, or maybe functioning at the tech level of AWS.

You just know that if the train in the East Palestine derailment was run not my Norfolk Southern, but by SBB Cargo, the Swiss national railways' cargo branch, then

1. The track would have been at least doubled, under wires, and secured using a very advanced standard of positive train control.
2. The train would have been several trains, each hauled by electric locomotives.
3. The disaster train, at best, would not even have made it out of the yard. At worst, it would have been stopped, and probably directed onto a siding, two towns prior for having a failing bearing.
4. Passenger trains would have all the room to run down the track they need.

The US freight rail industry isn't some of the best in the world, it's actually really quite terrible. It fails to maintain it's infrastructure, can't run to a schedule, frequently loses cargo, and causes ecological disasters. It is good at creating short term profits for shareholders, not being an effective transportation network. If you want more info, here's a video that explains it better.

It's also conveniently a bunch of companies that don't really overlap and manage effective monopolies in their region.

So what's the point of the OPs comment?

Fair response. It’s likely due to the lack of rail infrastructure why this delivery was “across the country”. Rails are typically much cheaper per ton-mile than trucks. If a rail alternative existed, I’m fairly certain the economics would have forced the use of trains.

However, I’d say the self driving part is still be a benefit that would improve truck utilization rate.

I did some digging. According to the article, the route went from Tulare, CA to Quakertown, PA. OpenRailwayMap is really good for this. Both have freight rail lines running directly through the heart of the town. Going by destination alone, this is kind of a pointless operation. Then again, the point was more to demonstrate the possibility of an autonomous truck rather than whether that particular route made any sense.

railways are a lot more expensive than roadways per km

compared to a 5 lane highway its a pittance - theres a reason why private rail companies can exist but private road companies largely don't.

The problem is there's a lot more federal funding for the shittier solution so when budgetting are you going to build the thing the feds will pay 100% or 0%?