Hacker News Reader

Comment by WalterBright | Hacker News Reader

I rarely see any talk about moving from long haul trucks to freight trains. The latter burn far less fuel, and steel wheels don't create synthetic rubber dust.

View on HN

This is a topic that is talked about in academia quite a bit (I have a master in supply chain management)

The main problem with trains is that you - naturally - have low flexibility in terms of where you can deliver goods. You are relying on trucks for the last-mile pick-up & delivery for the majority of trips. You might need to load/onload 4-6 times going from truck > train > truck. Because of the time & costs associated with switching mode of transport, these intermodal alternatives become uncompetitive compared to a single door-to-door delivery via truck, even for fairly long distances.

The logistics providers will optimize for lower cost over lower pollution, so intermodal transport via train can be hard to justify unless the customer is willing to pay for it in terms of money or time.

this is an example of how our economic choices are altered for the worse by negative externalities. we need to properly internalize the cost of pollution from fossil fuels, roads, tires, weight, miles, etc. so that economic choices are appropriately rationalized.

Trucks don't pay for much of their costs. For example, most of the maintenance of highways is the result of the heavy trucks slowly braking up the pavement.

>Trucks don't pay for much of their costs.

And IIRC companies like Linfox donate billions to both main political parties, which I'm sure has no effect on the decision to keep truck registration costs so low.

Do remember that damage to the road from tires scales exponentially with weight (per tire). So those trucks are doing huge damage to the roads (and not paying for it directly). Here's an article with a nice graph about it: https://streets.mn/2016/07/07/chart-of-the-day-vehicle-weigh...

> You might need to load/onload 4-6 times going from truck > train > truck.

That sounds strange to me. It implies that the US doesn't have a rail network, just rail lines that don't connect to anything.

Something gets made at a plant domestically. It gets brought to the train station to be shipped. Truck -> train. The train carries it a few states over to the wholesaler's mega warehouse. Train -> truck. Wholesaler breaks it down, groups it with other things to send out, Truck -> Train. Regional distribution center picks it up from the train station and takes it back to their distribution center. Train -> Truck. Then it gets put out as individual orders in the region. Truck -> Store.

If you're needing the truck to take it back and fourth from the train station, you may just find it easier and faster to just have the truck carry it all the way from the port to the mega warehouse, then to the distribution center, then to the actual delivery site.

The "few states over" is what should go on the train. It's the same idea as container shipping.

"Regional" here still encompasses a few states and several hundred miles for a trip. In the US "regional" distribution centers may cover the same geographic area as France+Germany+Belgium+Netherlands. Its not uncommon for New Mexico+Texas+Oklahoma+Arkansas+Lousiana to be one "region", if you're set up in Oklahoma or Texas each of those states is pretty much just one state over. Albuquerque to Memphis (the East-West stretch of this region) is 1,000mi.

My impression is that there’s a huge integration problem where rail freight is mostly set up to work with the sea-oriented container system, whereas truck LTL works on the pallet.

Or, framed another way: trucks are just too cheap. Truck operators and truck-oriented logistics firms don’t pay for their negative externalities, so there’s no cost benefit in going to the trouble of routing your load through a rail freight exchange on both ends. It’s pure complexity and schedule risk with no upside.

I don’t know where you live, but for instance in Argentina this conversation never happens because the union of truck drivers is so big that they stop every single attempt to increase trains.

We still very much use freight trains. But they don't go everywhere, and managing a network of train traffic has its own overhead. Each car in a train has to go to the same place. To transfer to another train to take a different sub-route takes time. Then, you also have to unload the payload at the destination and still complete the journey with a truck.

The load/unload should be as simple as picking up the semi (sans the cab) and putting it onto a flatcar.

It's never that simple. Freight trains are extremely long and have to have massive yards to park in. Then you have to shuffle them around to get to the car you need. Then you have to have the Semi Truck in a position that a crane can access to mount the payload. It's a very difficult problem to solve at scale and certainly not economical for every city in America.

This is not a complex problem, as the rail yard is a completely controlled environment. It is not a major engineering problem to have the train move slowly past a point where a crane can pick off the trailer and put it on a tug. The tug then parks the trailer somewhere in the yard. When the truck shows up, the tug retrieves the trailer and brings it to the truck.

Think of a moving assembly line. It's the same thing.

It doesn't have to be economical for every city. It can be done profitably in an incremental manner, starting with the highest volume lines.

Yes it is. Your long slow moving train is now blocking other long trains from passing (since rails are rarely two way). Nevermind that the train would have to stop entirely and lose all the energy efficiency in the process). Constructing new rails is expensive and difficult since rails have to be nearly level with very low grade.

You are handwaving about something you know nothing about and I suggest you read more into civil engineering, civil construction, and rail construction. Union Pacific will spend billions just to get roads reconstructed to move under or over train tracks so they can save a couple minutes on a delivery route. It's extremely difficult to build roads around or near freight rails because of these constraints and vice versa.

Have you even compared a map of the freight rail lines in existence and their current stopping points versus the interstate and US highway system?

They also don't leave PM10 floating in the city as much, by virtue of not being distributed widely throughout the city ubiquitously close to schools, hospitals, homes. So apart from just emitting less, they emit less, specifically close to where people are, compared to trucks

Warren Buffet talks about it every once in a while. He has talked about the efficiency gains in terms of how much it takes to move a ton of goods a mile. I have seen stories about what types of investments are needed in the Chicago areas to unblock bottlenecks between the East and West coasts.

This is very common for cargo movement in the USA.

https://en.wikipedia.org/wiki/Piggyback_(transportation)#Rai...

Trains can be electric.

In france almost all the main railways are electrics.

Reply on news.ycombinator.com