Railroads beware: Technology is gaining on you | Hacker News Reader

Railroads beware: Technology is gaining on you

railwayage.com

27 pointskposehn49 comments

Interesting thing about railroad stocks, and no doubt one of the reasons Buffet bought BNSF, is that they own one of the most valuable things in the world of transportation: rights of way. Is hyperloop the way of the future for freight? If so, the railroad companies can build that out faster than anyone else, because they've got the rights of way.

Incidentally, this is one of the things I don't get about Hyperloop versus CHSR: there is nothing inherently cheaper about building an evacuated tube versus putting down a few steel rails. You still need a right of way and grade-separated crossings, and you still need to do all the engineering/environmental work for all that and that's where the expense piles up. You think laying some steel track down on a rock bed is the expensive part? Now, once you have the right of way in place, that's going to be valuable no matter what technology comes down the pike. Amtrak estimates the cost of upgrading the northeast corridor to 220 mph HSR at $160 billion. A lot of money, to be sure, but imagine what it would cost if there wasn't that existing right of way from DC to Boston, through downtown Baltimore, Philadephia, and New York, already in place! That'd be a trillion dollar project, whether you did it with a HSR or Hyperloop or whatever.

I doubt Hyperloop is the future for freight, and freight is overwhelmingly where it's at for railroads (in terms of $$$). Hyperloop is built for passengers, where speed is the overriding concern. For freight rail, price of transportation is the primary concern. Accelerating a few tons of iron ore to nearly the speed of sound to get it somewhere is prrrobably not worth it.

I don't even think I really buy the "3D printing will change everything" take either. Railroads will still be perfectly useful for shipping 3D-printer feedstock to distribution points for feeding the 3D-printing micro-factories.

Unless the author says that passenger rail - in the US, that'd be Amtrak - should watch out for the future, in which case I'd say that's been patently obvious for longer than I've been alive . . .

The RoW is highly valuable for all sorts of stuff. Back in the day, iirc, Qwest was owned by Phil Anshutz, who also owned the Southern Pacific Railroad. He used the SP RoW to lay down huge tracts of Fiber, which the other railroads copied.

They can and do run all sorts of stuff long the tracks - fiber, natural gas pipelines, and soon high transmission power lines.

> Southern Pacific Communications decided they needed a new name to differentiate the switched voice service from SpeedFAX, and ran an internal contest to select a name. The winning entry was "SPRINT", an acronym for Southern Pacific Railroad Internal Network Telecommunications.

From: http://en.wikipedia.org/wiki/Sprint_Corporation

If you want to talk about "Back in the day", telegraph companies had been running their lines along ROWs since the 1860s. They would obviously have to duplicate ways across vast expanses of the country otherwise. In exchange, they would leave a few conductors open for the railroads to use for their communications.

The thing is that the speed of travel determines how quickly you can curve and that's built in to the right of way. The major limitation on the speeds on the current northeast corridor is the curvature of tracks which were built for slower trains. Without a link I can't be sure, but I believe that the $160 billion high-speed rail plan you're referring to required building new right of way, at least between New York and Boston.

There is nothing inherently cheaper about building rockets versus building rockets. You still need steel tubes, fuel, safety checking, failsafes, tons of programming, recovery tests, launch tests, etc etc etc. You think contracting out the launch site is the expensive part? Boeing and Lockheed, Bush, and Obama put a total pricetag of getting to Mars at ~100 billion. Why does Musk think he can do it in 10?

The major cost of building out a high speed rail line isn't even the right of way. It's that the entire structure of the system is geared towards making payouts to politically connected individuals, industries, and special interests. If you know anything about CHSR, you'd know that already ~$20MM has been spent on it without even laying down a single track (er, they might have started a smidge of work on it in an inconsequential section) or acquiring any rights of way. This is already way way way overbudget. If anything the overriding reason that Hyperloop will be cheaper is that it stands a chance of being a carte blanche idea - where the influence- and money- grubbing constituencies have at least a shot of minimizing their impact on the bottom line of the project.

Regardless of the merits of this argument, it's not relevant here. Musk didn't estimate the cost of the SF->LA hyperloop on the basis of "I have a unique ability to lower the costs of this project."

He estimated the cost as low based on the use of a preexisting right-of-way (mainly I-5), and needing an inherently narrower, lighter "track" than was envisioned for HSR.

A hyperloop designed for freight would presumably be wider than what Musk envisioned for the passenger-focused LA/SF route, and in some routes would involve either purchasing or repurposing rights of way.

> Musk didn't estimate the cost of the SF->LA hyperloop on the basis of "I have a unique ability to lower the costs of this project."

No, but he didn't bother to explicitly explain why the CHSR is expensive, he just gave their figure (which is almost certainly an underestimate). For example, unwritten went the parts about not servicing cities where there is less demand (mostly a political play by CHSR).... There won't be a hyperloop designed for freight - most freight runs perfectly fine on existing rail lines. Part of the reason why passenger rail in the US sucks is that it shares lines with freight the safety standards for freight are more lax, for obvious reasons, and the speed tolerances lower, so passenger rail is restricted to lower speeds on freight lines, then amtrak has to make exorbitant payouts to rent-seekers like BNSF and Norfolk Southern (the most egregious of them, IIRC) just to rent back that right of way.

for example, the Dianne Feinstein's husband's company chairs (CBRE) was awarded the contract for construction of segments of CHSR.

He's on the board of directors of CBRE. Also, CBRE is the preeminent commercial real estate company in the world. I'd be amazed if you could accomplish a project of that magnitude without involving that company at some level.

I actually made a mistake - it wasn't CBRE that won the award, but rather a company called "Tutor-Perini", which Richard Blum is the chairman of the board for.

There's all sorts of fishy things going on: http://www.utsandiego.com/news/2013/Apr/21/tp-high-speed-rai...

I doubt that 3D printing will be a problem for railroads. You still need to transport for the raw materials, and most people will not have industrial-scale 3D printers in their homes (any more than they have industrial-scale paper printers in their homes). There are also limits to what can be 3D printed, and those limits will remain in effect for the foreseeable future -- liquids, timber, food, and so forth are not things that will be easy to print, but they remain important and they are often transported by rail.

If anything American railroads should be rejoicing about 3D printing, as it will encourage more domestic manufacturing and create new opportunities in transporting both raw materials and finished products.

(Edit: I know it sounds like I am making the same argument people made in the 1960s about computers in the home. However, a key difference with 3D printing is that the size of the printer determines what can be printed. It is one thing to print forks, knives, and gears; it is quite another to print a car body.)

I don't think that you're making a "computer in the home" argument. It's more like tools.

In 1960, a typical handy homeowner had a hand tools. Sure, I could build a deck in the backyard, but it would take a long time and require more hands. Today, a typical handy homeowner has power drills, saws, etc at very low price points.

I was a bit dubious at first, but I understand what they're on about now:

The concern is that goods will be finished much closer to PoS, resulting in more transport of bulk commodities and less of finished goods. Intermodal transport (which includes those loooooooong double-stack trains you often see) is primarily finished goods going to market. If railroads want to stay competitive, I do agree that they need to "skate to where the puck is going" and make sure they have the capacity to handle shifts in traffic patterns.

A good example is the Powder River Basin. That is the largest coal producing region in the US and for about 30 years has received massive capital investment. However, coal is on the decline and crude-by-rail is on a massive upswing. The railroads got caught flat-footed there and have been scrambling to increase capacity for crude-by-rail.

As far as bulk crude, how else are you going to transport it outside of a pipeline? In this the railroads are safe, but in the future they may not be.

I'm also not sold on household 3D printers. Maintaining and operating a 3D printer would be a hassle. But retail-level printers? That's a no-brainer. I'm certain people would be willing to accept a reasonable mark-up to have things printed for them.

I'm seeing hardware stores that prints all the screws and nails on-site (or at a regional distributor, shipped just-in-time) rather than keeping inventory, and "here's a random part that broke on this obsolete product, could you replace it?" services.

I'm seeing furniture stores that create made-to-measure furniture of varying dimensions or colours from a standard template to fit your room. Like made-to-measure clothing, as opposed to bespoke (custom furniture) or off-the-rack (furniture store).

I'm seeing motorcycle/bike/hockey helmets made to fit a person's head, car seats moulded to a person's backside (currently only an option on super cars like the McLaren F1), and custom orthotics that are cheap enough to become mainstream.

I'm seeing mall photo booths that produce figurines instead of photos.

But no, I can't see people having a 3D printer in their basement to do any of these things in their house.

I can see it but only with the technology changing. My husband has a 3D printer and uses it all the time for little projects around the house. Once the technology advances, you can have printers that are the right size and the right interface for the common public, they can 'buy' or create a design and have it just automatically print. Think of having toys for kids or household products printed faster than driving to a store to get it. Right now 'makers' are already doing this. It's not very friendly for most people, but its been very helpful for household uses.

I had a hard time reading this article without wondering if it was some kind of spoof. The USA has the most efficient freight rail network in the world (other countries optimise their networks for passenger), and this isn't due to a single factor. It is never going to be under threat from the Hyperloop, because most of the cargo carried are things like coal, raw materials, containers, etc. that aren't sensitive to being delivered "just a few hours faster" (or even days). Also, most of the US freight rail network is heavily heavily optimised and drops containers off for truck haulage after transporting the cargo to the most efficient spot on the network.

I heard about how the modern diesel-electric trains are engineered and wondered what I could learn from their systems, and was just about shocked. My intent was a thought experiment on "could we build a better hybrid car?" And the answer is a resounding yes... just look at trains.

All the research is already done, by train builders. They've done the math that says when to prioritize amperage over voltage for efficiency, and how to optimize various systems. And that's just one thing that I found massively impressive. (Now if only I could find the site I read, it was impressive.)

Find it! I'd really like to know what you found.

I've done a fair amount of research myself and have always felt the auto industry could learn a lot from locomotive builders, especially the use of 3-phase ac traction.

One caveat: there is a ton of highly time sensitive traffic, especially FedEx and UPS z-trains that run on very tight schedules - and the railroads charge accordingly.

By volume and especially by weight, time sensitive mail deliveries are a tiny, tiny fraction of the bulk of coal, grain, scrap metal, etc. that make up most freight traffic.

True, though it is highly profitable as well.

Hypothetically, wouldn't a high speed freight network coupled with just-in-time production, and a amazon-like online shopping with one day delivery change the face of retail industry.

Railroads are much more valuable than most Americans can possibly imagine. And technology will only make it more so, if sufficient people are willing to work on it. Here are some ideas off the top of my head:

* automated trains for long distance travel, self-driving car-2-go's for short distance travel

* integrate future transport modules with the railway network and build more tracks; travel long distance on electric self-driving cars.

Of course, the main obstacle is the high initial capital investment. Here too, anyone really serious should start small: start by helping the current rail operators optimize freight (the real money maker); gain their confidence and then work on a joint project on a small scale; if it works out, then expand throughout the country.

It's painful to see that a country as large as America has such an inadequate railroad system. But maybe it was for the best: countries like France or Japan with an entrenched railway system will have probably have a hard time making the switch to radical new technologies

The main obstacle in the U.S. is that the interstate passenger rail system was nationalized by Congress. Congress forces unprofitable long-distance routes on Amtrak and keeps the monopoly company undercapitalized. Amtrak owns amazing assets like the Northeast corridor rail lines, but there's almost no innovation as the monopoly loses billions of dollars and can't really tap private capital markets.

Meanwhile American freight rail-- totally privately owned and managed-- is the best in the world.

To be fair it was nationalized in the wake of the Penn Central bankruptcy and all of the other railroads wanting to drop passenger rail. Railroads were even able to opt out and I think a few did for a few years but it just didn't make sense to move passengers vs freight.

The actual problem was the federal Interstate Commerce Commission regulated routes and ticket prices and wouldn't give passenger rail companies the flexibility needed to compete with cars and growing air travel. Central planning failed and completely destroyed U.S. passenger rail.

Could such a system, on a more grand scale, replace railroads as conveyors of passengers and freight? The vision is not new-fangled, and a modern day visionary insists it could be a better alternative to high speed passenger rail.

The interesting thing is that the incumbent railroads are actually in a prime position to capitalize on this tech. The hard part about implementing the system isn't really the technology, it's all of the property, right-of-way, and NIMBY issues that come up. However the railroads already own property all over the country and (for the most part) they can do what they want with it.

Everyone might be better off working together than trying to approach it separately.

edit: beaten by rayiner

Driverless trucks could also be a threat/opportunity. These trucks will trundle along at the most fuel-efficient speed 24 hours/day. Eventually they may be able hook together on the highway, further reducing drag.

Of course trains can also become driverless and train-cars could automatically couple/decouple, which could lead to many more trains, some with much smaller loads. I can even imagine trucks someday with wheels for both highway and steel-rails.

I think that driverless trucks are a natural migration of the short-haul model, where you have a container that hops off a train at an intermodal facility and is driven to the customer.

Automated trains are definitely coming; look at BHP Billiton's push toward remote train operation in the Pilbara region of West Australia.

Interesting long-term issue.

I think the next decade or two of railroad history are all about moving commodities like oil and produce. Hyperloop is all about passengers (something the rail industry has never cared out) 3D printing requires raw materials to print stuff, which will drive more demand for rail transport. Nothing can move billions of plastic pellets faster and cheaper than a train.

Well, we may find a hyperloop is better for transporting the smaller pieces of a commodity. After all, if you can just send a crate instead of a container, you can send to the buyer exactly what they need and nothing more. The issue with bulk freight is once you've made the pellets, at some point they have to be divided up.

I think something like a hyperloop will do very well for less-than-carload/container freight, but would probably never transport the massive volumes of bulk products (chemicals, wood, oil, coal, ore, steel coils, etc.)

>I think something like a hyperloop will do very well for less-than-carload/container freight, but would probably never transport the massive volumes of bulk products (chemicals, wood, oil, coal, ore, steel coils, etc.)

Exactly. The hyperloop network would probably cost much more due to the extensive passenger safety features that need to be (hopefully) built into it

Railroads lost 99.9% of that business 70-100 years ago.

You ship stuff that doesn't have time-definite requirements with UPS Ground or Parcel Post. Time definite, you use an express service. If you have time definate needs and larger quantities, you use FedEx Custom Critical, or some LTL truck service.

Some folks today ship same-day parcel shipments via Amtrak, Southwest Airlines, or other passenger service. That's what hyperloop parcel delivery would compete with. It's onerous and expensive (ie. I need a guy to drive to the station with the thing I need to ship), and only makes sense if you have an office adjacent to the city pairs or airport. Lawyers used this service alot pre-internet.

Who took that business in 1913ish? USPS was carried via rail until at least the 50s.

Trucks powered by internal combustion engines? It's not an all-or-nothing thing; many products even today have a rail segment somewhere in their delivery to the consumer. Even so, for rail to grow much it has to take business from trucking.

There wasn't much in roads or trucks for long distance travel in the early 1900s. I believe it took roughly 30 days to do the trip across the country on the Lincoln Highway as opposed to 7 days via the train.

I would expect if you look at the volume of freight shipped via rail vs trucks it wouldn't be until the 1950s and the creation of the interstate system that it really took off. Containerization in the 50s and 60s would be another big boon to trucking.

Do you think that a hyperloop wouldn't be practical for high-margin finished goods?

I'm thinking things that are intermodal freight, like a container of TV's bound for a CostCo in San Diego, shipped from Long Beach.

> Hyperloop is all about passengers (something the rail industry has never cared out)

This is disingenuous. In the 150+ years of North American rail, many of America's most intricate architectural structures[1] are train stations, and they certainly weren't making that investment for freight. The dining car service was also typically a loss leader, designed with the sole purpose of attracting customers. Freight never complains, though, and ends up being much better to transport.

[1]: http://www.nps.gov/stea/planyourvisit/lackawanna-station-100...

There is about ~121,800 miles of track in North America owned by CSX, Norfolk Southern, Union Pacific, BNSF, and Canadian National.

That's about ~4.9x the circumference of the earth.

Seems like it would be quite expensive to build this new infrastructure? Maybe if highway transportation was more fuel efficient / fast the amount of tubing could be minimized.

In contrast, Germany, a country that is roughly 1/20 the size of the US, has about 60000 kilometers of track, or about 1.5x the circumference of the earth.

The hyperloop would actually be great for railroads to get involved in. They already have plenty of straight-line right-of-ways, they have deep pockets to finance the tech, and with the right corridor they could get into the same-day delivery business. Perhaps even be a direct competitor to Fed-Ex.

I'm not sure the industry is ready to take those kinds of risks yet, though. The entire hyperloop concept is great, but it's just an idea, and ideas without any execution intelligence don't tend to do so well.

"If you made it to age 50, you recall department stores where clerks placed a sales slip and cash in a cylindrical container that they then inserted into a steel tube at the point of sale, with the container then transported by compressed air and partial vacuum, through the maze of tubing, to cashiers elsewhere in the store."

Many supermarkets in the UK have these at the till.

I have never seen one in the UK... where?

lots of the bigger ones - http://www.telegraph.co.uk/news/uknews/1554632/Thief-stole-9...

Ah interesting, makes sense I suppose to get cash away from tills.

Take note that Musk's Hyperloop prototype was created on a 3-D printer,

What prototype? Did Musk actually show a model?

That's far more interesting than the title would imply.

Reply on news.ycombinator.com