Acela Replacement and Disposition Discussion

we really think that there cannot be any prediction of 5 year out ridership when AC-2s are in service.. Anything from about the same as today to twice today. How many persons can be attracted from the roadway traffic has too many variables.

It has a number of game-changing variables, like self-driving luxury cars, but the Acela II is also a game-changer. In the end, I-95 will still be I-95, DCA and LGA will still be their respective selves with security theater, and those who can afford it will still prefer to take a train between NEC points (even if they can’t always do so.)

A little OT -- will not be offended if moderator makes it a new thread -- but related: If self-driving cars make taxis and zip-car-type-things cheaper and more convenient, that could mean fewer people going car free, maybe even in smaller cities that don't have great public transportation. People who don't own cars -- who aren't already automatically paying $500 or $1000 a month in loan pmts, insurance, etc. -- are more likely to ride trains because it makes more economic sense for them to do so: if you own a car, the cost of a train ticket and maybe rental car duplicates money you can't avoid spending, but if you don't have a car, you compare the cost of train and rental car etc to the cost of renting a car and driving the whole way. Train is more likely to win in the second case. That is why a place like the Netherlands can have good load factors on intercity trains running everywhere at half-hour intervals or better behind subway-style fare gates but with generally good seats and fast schedules. Even BosWash isn't as dense as the Netherlands, except in a fairly narrow corridor, and Netherlands has its many cities in a square area so none are more than about four hours apart, so it may not be the best comparison, but the basic point holds: If self-driving cars convince more people to stop owning cars, they will almost certainly raise public transit ridership. Heck, even if they just make it a little more reliable to get a cab in a hurry if you miss the last train or need to make an odd transverse trip that doesn't work well on public transit they will increase public transit ridership. And with that they will probably increase intercity train trips, too.

I can't promise that self-driving cars actually will lead to fewer people owning their own car, but if they are more expensive than the cars we have now, but cheaper to operate as cabs becaue they don't have drivers to pay, then they will bring the cost of owning a car and the cost of taking cabs when you want to closer together, and that would probably tend to lead to more people skipping the ownership and using mass transit when convenient because cheap, and cab or zip when needed because it is fast and convenient.

Most of this is probably moot in most of the US because the distances between cities are so much greater and the population density so much less that it would take a lot more change in urban settlement and daily travel habits to move people away from owning their own cars. And there are various wrinkles: someone like me, who wants to be able to go hiking nearby often, and a few hours away from home a couple times a year, may think a car is more important than a neighbor who mostly enjoys computer games or the local gym or whatever (even though I don't really like cars, I like getting to trailheads. Here in Oslo, there are a bunch of buses to trailheads with 20-mile trails straight out into the woods, but not many cities are like that, and I am going back to the US in a few months). However, in the NEC area, greater Chicago, San Francisco-Oakland, and a few other places, there's a reasonable chance.

And it is certainly possible that Amtrak could gradually build up NEC capacity by using longer and faster trains and making incremental track and signal improvements that would in tun allow for greater traffic density and higher speed, so that eventually someone at Penn Station NY would have a choice of two or three classes of train (express, regular, commuter) to Boston, Spg, Wash, Albany, Harrisburg, etc., with no more than a 15-minute wait for a decently fast, decently comfortable ride to most NEC destinations, with walk-up walk-on unreserved rarely full service for all but the long-distance trains. Somewhere in there, of course, is getting a national consensus to fund the big capital improvements, or a strong state coalition to do the same. Back in the 20s the Mississippi Valley congressional delegations and the Colorado River delegations traded votes, so that the westerners voted for midwestern flood control and the midwesterners and southerners voted for western dams and irrigation. Before that, a majority of congress dismissed each of those as wasteful special interests. What could the urban/dense part of the country trade to the rural/sparse parts of the country so that both would get something they value? Right now it often seems like we (I am Bostonian at heart) want everyone to ride the subway and they want everyone to drive, but it is really more that different transportation mixes make sense in different places. Or maybe what I mean is that it takes a lot of traffic jams for many people who can afford a car not to want to drive around, so most of the country sees mass transit as an obsolete, slow, inconvenient thing for poor people and rail fans and environmental nuts (I am the last two, but not all passengers are).

Arborwayfan wrote:And it is certainly possible that Amtrak could gradually build up NEC capacity by using longer and faster trains and making incremental track and signal improvements that would in tun allow for greater traffic density and higher speed, so that eventually someone at Penn Station NY would have a choice of two or three classes of train (express, regular, commuter) to Boston, Spg, Wash, Albany, Harrisburg, etc., with no more than a 15-minute wait for a decently fast, decently comfortable ride to most NEC destinations, with walk-up walk-on unreserved rarely full service for all but the long-distance trains.

Right now it often seems like we (I am Bostonian at heart) want everyone to ride the subway and they want everyone to drive, but it is really more that different transportation mixes make sense in different places. Or maybe what I mean is that it takes a lot of traffic jams for many people who can afford a car not to want to drive around, so most of the country sees mass transit as an obsolete, slow, inconvenient thing for poor people and rail fans and environmental nuts (I am the last two, but not all passengers are).

Where there is density of both people and traffic jams, we already have a choice of classes of mass transit (why limit it just to trains). Commuter and urban metros already exist alongside Amtrak's intercity train services, plus Amtrak's high speed Acela services to add another class to intercity train services. But that density is localized into specific cities or metros.

Taking your example of the Netherlands, it's relatively small in area not only compared to the whole USA, but to the NEC supermetro.
Per Wiki, the area of the Netherlands is 16,040 square miles.
Here's how it compares to various NEC States:
D.C. 68 square miles
Maryland 12,407 square miles
Delaware 1,982 square miles
Grouping subtotal approximately the same at 14,457 square miles.

Pennsylvania 46,055 square miles (individually almost triple as the Netherlands)
New Jersey 8,722 square miles (individually about half as the Netherlands)
New York 54,555 square miles (individually more than triple as the Netherlands)

Connecticut 5,567 square miles
Rhode Island 1,214 square miles
Massachusetts 10,565 square miles
Grouping subtotal approximately the same at 17,346 square miles

Grouping all the NEC states together for a total of 141,135 square miles, they would be 8.8 times larger than the Netherlands in area.

Additionally, the Netherlands has basically one major powerhouse city of Amsterdam, the NEC has four; Boston, New York, Philadelphia, and D.C. The NEC has seven commuter rail operators servicing those four major powerhouse cities; MBTA, CTA, MTA North, LIRR, NJT, SEPTA, and MARC. We could even throw in an eight commuter rail operator VRE. And we're not even counting the urban metro and subway systems yet.

Let me repeat myself, where there is both density in people and traffic jams, there's very frequent service along the NEC in the USA. Where there isn't, there might be no rail services at all. Not every town in the Netherlands has rail services either.

electricon, I think you may be violently agreeing with me about the key points of a comparison to the Netherlands. More or less. It is a small place, with one huge, several large, and several midsized cities spread out over a grid as opposed to a line. Amsterdam is not much more of a hub than Rotterdam, for example, and many routes involve a change at a relatively small city (Breda, Deventer, Den Bosch, Utrecht...). The area that is either in a dense city or not too far outside a dense city (local bus distance) is, I think, larger than any similar area in the US. A lot of Delaware is really outside urban areas, and so is a lot of Maryland. The whole southern two thirds of the country, at least, function almost like a metropolitan area. Where there aren't trains, there are buses. There is really no place in the US developed in that shape, with that strong of a bike bus and train habit, or -- I insist -- that frequency of buses and trains over substantial distances. At the level of service the Dutch have, there would be 15-minute service on the whole NEC and the Inland Route and to Albany and to Harrisburg and to a slew of other destinations we can't think of. The gap between the T and the Shore Line East would not exist. Their network is very dense and very busy, and it matches their development.

WE are still very far away from self driving cars,Both Tesla and Uber had some BAD failures in the recent past.
Most notable: Uber self driver with onboard pilot vs a pedestrian in Phoenix Az.

Right now Autonomous Vehicle (AV) crashes are like airplane crashes: they are big news, but statistically ever-rarer compared to miles traveled, and the newsy-ness of their failures is masking their true level of safety and continuous progress. And it is an open question if any driver could have avoided that Phoenix crash (I'm conflicted on what I've read).

I'm pretty sure Tesla is already better than the median driver in terms of accidents per mile driven, and Waymo is considerably better (just not yet for general sale) and the best AV system will be pretty much winner-take-all (the 3rd or 4th best/safest AVs-- maybe that's Uber--will quickly lose market share to the better AV system, just like the 3th and 4th best search engines and word processors did). And there are ever better sensors to underpin whichever system(s) win(s).

For trips versus Acela, the car doesn't have to be better on 100% of the point-to-point trip, it just needs to fully relieve the driver on 3hr, 4hr, or 8hr highway stretches on the competitive interstates (95-direct, or all the alternatives on 81/83, 78/80, 284/84/90). For most of my NEC-parallel trips, I have about 10 miles of non-highway driving, that are no more of a burden than somebody driving me to South Station. My NEC trips involve 200 to 500 miles of interstate driving that are the bulk of the burden and would become the bulk of the pleasure of going by AV.

The competitive solution for Acela is higher speeds, and let the buses bear the brunt of AV's success.

Self driving cars are likely to slow down the interstate highway system one way or another - by being so popular that the interstates become more overloaded and/or by just driving conservatively and observing the speed limits. It is unlikely that auto autos will be permitted to program for what is currently illegal behavior, like tailgating or speeding, even if they are fairly common. If all vehicles were to maintain "safe" spacing between themselves on the interstates, the interstates' capacity would overnight become inadequate for even the current loads, let alone increased usage. If regulators and/or car people allow for auto autos to exceed real-person speed limits or intentionally program them to drive closer to other cars than the minimum distance necessary to stop, they would be crucified in court once that came out in the first big liability case. They hope to be using tech to prevent that sort of running too close together thing (which of course is just the opposite of allowing it on purpose) when regulating train speeds. even with the trains' vastly more controlled environment. Not that the gov doesn't sometimes do dumb or contradictory things sometimes, but I would be surprised to see them allowing autos to be intentionally driven unsafely while at the same time forcing train operators to always maintain safe spacing. The'd at least have to fund a study showing that unsafe was safe, or some such . . .

We've strayed far off topic, but I have to add to the off topic discussion just a little. Most states suggest 2 seconds between automobiles in the same lane of travel as safe, therefore each lane should only be able to safely handle 1800 automobiles per hour.
Math = 3600 seconds in an hour, 3600 / 2 = 1800.

Self driving cars on non-fixed guideways are one thing. The need for the ability for a human driver to take control when needed will still be there IMO.

Self-driving trains on fixed sets of tracks with 2 rails is another and very doable. I don't think/know if the technology is there yet for standard passenger rail operations, but it has been on the table in the past for rapid transit systems (BART and the Montreal REM light rail come to mind). BART was in fact originally conceived as being fully automated; it was later deemed to be better for there to be an operator for the benefit of the general public feeling safer having someone in the cab.

andrewjw wrote:

silverliner266 wrote: I was assuming they would be limited to runs in electrified territory. Unless I'm reading the timetable wrong, a decent number of trains only do WAS to NYP or BOS.

Currently, yes. However, once the Acela 2 is present, the plan is to increase the frequency of Acela services. I expect that some Northeast Regional services will be cut due to being redundant to an Acela (2) train. So the number of trains within the WAS-BOS segment with no low-level stops may become minimal.

I doubt Regionals will be cut back. They're too popular with customers on smaller budgets so -though I don't see an increasing market for them (maybe I'm wrong)- I think we'll see service retained at similar levels to now. NYP-BOS service for example will see no reduction in Regionals but will see a 5 round trip increase in Acela service through at least 2030.

Definitely getting off topic, but it is a great topic worth it's own thread.

Faster and/or cheaper and/or more convenient. That's the key to winning "disruptive" transportation solutions. Acela II will be faster & cheaper. Self-driving cars will be cheaper & more convenient. Stage coaches were cheaper & more convenient than a horse carriage. Steam trains were faster & cheaper than a stage coach. Streetcars were cheaper & more convenient than steam trains.

As for running vehicles closer together, you can have their computers talk to each other & "dock" while in motion.

Self-driving cars Achilles' heal could be inability to make eye contact or read body language.

electricron wrote:Additionally, the Netherlands has basically one major powerhouse city of Amsterdam, the NEC has four; Boston, New York, Philadelphia, and D.C.

What about Baltimore?

I'd put Stamford(/Bridgeport/Norwalk, same MSA as NYC) and Newark (almost far from NYC as Baltimore is from DC, technically the same MSA) before Baltimore.

I left some of the stuff that at least tangentially related to rail vis-a-vis competition. But Tadman is most correct; we're way afield.

Further mentions of self-driving cars will be banished to elonmusk.com