My father the transportation expert points out something else I should have said about freight rail: the railroads don't want to take freight that's going less than 1,000 miles because it isn't cost-effective for them. If the average truck trip is 700 miles, there's a huge gap that needs to be filled before freight rail becomes a meaningful replacement for a lot of truck trips.
He also notes that one of the giant problems with rail yards is that cities like to rezone around them. This severely restricts the usage of the yards, because the neighbors complain if you operate them 24/7. This is putting even more pressure on the rail yard system.
Most of the cost of rail shipment is the fixed cost of loading the car and hitching it into a train.
The real problem is railways in its present form were invented when engines are costly and hence,have to be built large in its own 'dedicated' car.
As modern electric trains show, there is no reason why the engine cannot be decentralized so each car is individually powered.
It is not a big leap to go from an individually powered car to an individually piloted car each with its own computer.
On top of this, if rail cars were built of modern materials (composites) and not have to bear the stress of being in a train --- it needs to be strong enough to withstand the pulling and pushing forces), it can be made much lighter and more energy efficient.
If you do all this, automated train pods can be made to negotiate much steeper grades and tighter turns, and / or travel much faster.
A new from the ground up train built like this can be much quieter, and also be routed through more neighborhoods. There is no reason to have noisy rail yards.
That is why trains as it stands are part of the problem. They are an obsolete technology.
To help her understand the latter point, Jane Galt can, for instance, spend a couple days in a motel on Route66 in Flag. Even a couple miles away from the tracks they're extremely loud at all hours of the day, although my experience might be a few years out of date. Maybe she could join MattY on his next excursion Out West.
How much of the expense involved in loading and assembling trains is labor? I'd assume a huge part. and how much of that labor is way overpriced because of unionization? Compared to, say, the costs of maintaining the infrastructure. Trucking gets their infrastructure costs spread out over passenger motorists.
A shakeup in rail labor costs could make rail competitive again for shorter hauls or for other market segments (passenger?).
D's comments about the engineered-in legacy costs of the technology are intriguing but won't change the labor costs.
heh, depending on where Megan went to school when she was in Chicago, she may know the sound of rail yards quite well. "They come by so often you wont notice" doesn't just apply to EL's...
As for re-doing the tech because it's obsolete.. that is where our rail density is totally against us. I have seen some info that we beat Europe in fraight movement via rail by 38% to 8% for some very interesting reasons:
Freight Share
so, oddly, we are actually doing better than I knew...
perhaps Megan's Da knows, but I couldn't find a good figure on how many freight cars are in the US, and that is where we have an issue. Changing them all to be autonomous self propelled, and light weight, is potentially logistically impossible. Who would build so many? They would have to work with the current infrastructure, to do it gradually. That infra is really oldstyle.
Maybe working with rail isn't the best upgrade.
On top of this, if rail cars were built of modern materials (composites) and not have to bear the stress of being in a train --- it needs to be strong enough to withstand the pulling and pushing forces), it can be made much lighter and more energy efficient.
At this point, you've just designed most of a long-haul truck, except that the truck can share right-of-way with commuter vehicles on existing roads, and access the destination point directly.
"My father the transportation expert points out something else I should have said about freight rail: the railroads don't want to take freight that's going less than 1,000 miles because it isn't cost-effective for them. If the average truck trip is 700 miles, there's a huge gap that needs to be filled before freight rail becomes a meaningful replacement for a lot of truck trips."
Surely cost-effectiveness depends on price? The huge advantage that rail has lies in energy costs, which is only going to become a bigger advantage over time. That gap between 700 and 1000 miles could be filled by a sufficient increase in the cost of energy alone.
"How much of the expense involved in loading and assembling trains is labor? I'd assume a huge part. and how much of that labor is way overpriced because of unionization?"
Big assumptions, and not justified by any citations. How much of the expense of trucking is paid for by the public in the form of subsidies for road building/maintainence?
There is a catch, mouse!
Long haul trucks run on diesels through a drive train that ends with rubber tires on pavement.
This system is inherently less efficient than trains that run on rails, which has much lower friction.
The efficiency of diesels are actually quite good, but they require liquid fuel which is a big issue. Electricity, though costlier to install, can be generated from nuclear, coal, etc. while there is no serious replacement for diesel on a regular truck.
If you run the numbers with a decent simulator, you will find that on an system wide energy efficiency basis, you are better off to have light / medium duty rail. There are some gains from regular freight rail, but costs as well.
Oh as for labor costs.... I am assuming that there is no way to de unionize existing unionized railways. A good part of the labor cost is also because trains / freight cars are "dumb" and need people around to marshall them in a yard.
Use smart cars that couple themselves, propel themselves individually around a yard, and also semi-automate the loading, and the numbers look quite different.
Just like trucking. Right now, a good chunk of the cost is the driver. Take the driver out, and you have eliminated the 2nd largest cost after diesel @ $4 a gallon.
This system is inherently less efficient than trains that run on rails, which has much lower friction.
This I am aware of, but by the time all of the rails and supporting hardware are installed and the necesssary right-of-ways are effectively closed off to any other possible use, plus accounting for the scheduling issues associated with negotiating use of the rails, plus the need to ferry cargo to and from the rail siding, the cost and lessened resource intensity of rail is not at all clear.
Most of the potential cost savings that could legitimately be associated with this scheme are already achieved via air freight (for lighter long-distance shipments) or conventional rail freight (for heavier long-distance shipments).
all of the rails and supporting hardware
Keeping in mind that supporting hardware, in the context of the present proposal, means more than 140,000 miles of electric grid and the associated power supply infrastructure, plus intermodal yards to bring the fancy 'lectric cars down old-fashioned non-electrified rail lines.
This is not a minor upgrade for improved efficiency.
This comment deleted for exceedingly poor taste in choice of targets.
Comment deleted for annoying the hell out of me. It is no secret that my father was the head of a trade association, but I don't allow commenters to insult my family members.
Oh, see, and just when I was about to respond to CB and tell him he was setting up a false dichotomy.
Well mouse, that is why I am talking about architecturing a system from the ground up.
See previous post on Freight Rail...
BTW, by no means I am underestimating the cost of an entirely new transport infrastructure.
Look at the present one: petroleum that need to be discovered, extracted, transported, refined, then distributed via terminals, filling stations, etc. to power virtually everything that moves.
The auto based one that need filling stations, roads, motels, rest stops, bridges, etc.
Nothing comes for free --- the problem is how we make better use of energy, particularly liquid fuel that is becoming scarcer all the time.
Transport by water is the best deal energy wise, followed by rail (even in its present form), more advanced forms of rail, and then trucks, then cars.
Nothing beats the good old bicycle, but there are many things a bike cannot do.
Well mouse, that is why I am talking about architecturing a system from the ground up.
Unfortunately, propositions that require an all-or-nothing commitment tend to die in the cradle, because the 'all' option is almost useless during the installation phase and causes enormous disruption of existing infrastructre for years while that installation takes place.
The reason railways never conceived a range of sizing options between "passenger trolley" and "Union Pacific" is precisely because the infrastructure costs, right-of-way requirements, and grading limits simply do not scale very well in between the extremes unless you have an application where other options are equally difficult to install and more time-consuming to use, e.g. a high-speed maglev passenger monorail across mountainous terrain beats a roadway every time if there is a need for passenger travel through that area.
Squeak, squeak.
I concur that the system need to be built bit by bit rather than as a mega project.
What might be the solution would be a set of interoperability standards but a lot of leeway given to individual operators to build their own trains as long as it is inter operable.
Insofar as right-of-way, that may be one of the easiest problems to solve --- existing interstate and highways that will become redundant as liquid fuel cost rise sharply, causing an absolute decrease in the amount of vehicles on the road.
If the system is architectured from the ground up to replace the tractor trailer and car, one of the advantages is that existing roads - with bridges pre-built - can be taken over. That assumes that the new medium duty rail can handle the grade involved --- not that big a problem.
I concur that the system need to be built bit by bit
You've got a huge network effect here, though; it's only really useful and efficient if there are lots of miles of track, or else you eat up all advantages at the intermodal terminals. Likewise, if you intend to cannibalize existing roads, you're talking about making dramatic reductions in the utility of the interstates to car travelers with no corresponding increase from the new rail.
And of course, I would expect the government (the only party that can credibly fund this affair) to bail on the project at precisely the point where the usefulness of the whole system has been optimized to lie at a global minimum.
Apparently all those hours your old man spent trapped in a room with Matt Rose working on that STC report were well spent. He is exactly right on the issues of long vs. short haul for railroads. The issue of encroachment limiting our expansion is also one of the thornier, and most under-appreciated, challenges we face. Seattle, Denver, Los Angeles, Portland, Fort Worth, the list goes on of places where folks build expensive houses right off a main line that sees 50 trains/day, and then start lighting up the phone banks with complaints. Try getting a line or yard expansion permitted next to a development with thousands of McMansions in it.
The main reason more traffic does not move on the railroad is because the service, reliability, and capacity of the rail network falls far short of its potential. The reason for that is insufficient investment, both in maintenance, capacity, control systems, and equipment. This is not to fault our executives, however, they simply cannot approve investments that don't pencil out with a positive NPV, and they have to be extremely disciplined about budget management. It never ceases to amaze me how efficiently and safely our Transportation, Engineering, and Mechanical Departments manage to keep everything running on the shoestring budgets they have.
Railroading is one of the most capital intensive industries in the the US, but government intervention keeps it from earning enough money to justify the level of investment it needs. For industrial and consumer products we compete against a subsidized trucking industry. If trucks paid their share of highway costs, railroads would have the top-line room to afford to compete in much shorter lanes. Where trucks cannot compete against us on cost and service is not as important, such as in hauling bulk commodities like grain, and coal, government action constrains our pricing to ridiculous levels. Inter-railroad competition is also pretty significant, regardless of what everyone says, but I do not object to that.
These price constraints keep us for earning a sufficient level of profit to justify the massive investments that we need to make in our infrastructure. Frankly, railroads are being forced to subsidize their shippers. Real rail rates have fallen in the last 20-30 years, even as service and capacity have improved. However, those days are gone - the rail networks are now full. There is not more room to shove trains through without significant investment.
I am afraid to say that the existing Railroad system is not much good for anything else beside what they are already doing --- hauling mostly low valued, non time sensitive freight like coal, ores, agricultural products, etc.
It would be a mistake to mix that type of cargo with a future transportation system that is intended to carry high value, time sensitive freight or passengers.
Whats more, the people that run the existing system are too treaded to the old way, and what is needed is new young minds, new thinking, and to jettison the old railway habits and thinking. Without changing the software, so to speak, the new hardware cannot do much good.
Look at all the examples of railroad based mass transit: NJ Transit, Path Trains, or most city subway systems and you see an iron age technology at its finest. We can do better than this in the age of smart machines.
So I am afraid to say, good olde grumps... you will have to grump at trainloads of coal or whatever is left of Amtrak.
As for taking room from existing roadways.... 90% of sub- and ex-urban and rural roadways have plenty of room for expansion either alongside or in unused portions like the center median strip.
So before we talk about reducing capacity (which might be the case at certain choke points, lets push this thing a bit further first.
It is not necessary to build a large system from the get go --- go for the low hanging fruit like sub- and ex- urban areas which is primarily serviced by cars and light/medium duty trucks where the break-bulk has already been done. Worry about connecting these nodes to a national network or expanding them to an urban area later --- that is why the importance is to emphasize interoperability.
It would be worthwhile to do some studies to see what sort of costs per mile we are talking about, or costs per ton km / cubic meter km we are talking about.
The economics are not there yet. It will need gasoline to hit $10 a gallon and diesel to hit $12 to $15 for a fundamental rethink of transportation.
That is not likely to happen before 2015. But come to think of it, given the lead times involved, that is just around the corner.
D: "multiple unit freight is gonna save us all!" has been around for a long time, but it's still not a good idea. You're focusing far too much on acceleration, which is not much of an issue in freight trains.
Passenger trains are light weight machines. Thus, a lighter train makes a big difference in total weight, and due to the number of start/stop cycles weight makes a big difference. That's why the Underground is MU. Those constraints fade to nothing when you're talking about hauling a hundred open-tops full of coal across Texas non-stop.
Most of the advantages on the systems side come simply from ICE efficiency increasing as the system is scaled, and on the operational side from the train not stopping.
The vast, vast majority of the time a train is running at steady state, non-accelerating, with relatively low driven wheel traction loading, and thus low slip. Csf on the surface is irrelevant because you're getting nowhere near the breakaway point anyways. We're not talking rally cars here. The big drag is aerodynamics and internal bearing/friction loads.
Sure, I think there's a lot of utility in semi-autonomous DMUs - but that's mostly for routing reasons.