Bonas Stop Comparing Amtrak ridership with automobiles. You have to use % of "Common Carriers" that charge a fare for a seat like the Airlines and Greyhound.
Stop Comparing Amtrak ridership with automobiles. You have to use % of "Common Carriers" that charge a fare for a seat like the Airlines and Greyhound.
Amtrak competes for the traveler's intercity dollar irrespective of the mode. In FY10 (the latest numbers crunched by the U.S. Department of Transportation), passenger rail (mostly Amtrak) accounted for 13/100s of one per cent of the intercity passenger miles. Buses, including transit running more than 50 miles, racked up 1.08 per cent of the miles, and commercial air carriers rang up 11.6 per cent. The car came in at a whooping 87.18 per cent of the passenger miles.
The biggest factor that contributed to the demise of the passenger train was the car. If intercity passenger rail is to have a viable future, it will be predicated in part on the ability to pry people out of their cars and onto the train.
BonasThe bus goes places that Amtrak does not go.
But Amtrak goes places buses don't go. And buses are going to fewer and fewer places these days. The discount operators make their profit by skimming the heavily traveled routes and ignore the people in smaller places.
Like the Idea of City Pairs,,,The bus goes places that Amtrak does not go. College Students and Leisure Travelers make up a huge portion of Amtrak Passengers. People who are not sent by there companys on some errant mission but choose to ride Amtrak
schlimmUnless that destination is in the center city or close to a commuter rail or transit line, a car rental may be required.
Which, if you are under 25 years old, is impossible and/or pricey.
-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/)
Correction noted. I was using "sustained" as used in retailing to refer to the sustained or average markup over the life of an object in stock and sold. You touch on another issue for passenger rail travel in the US. because of our auto-dependent urban development compared to Europe's, taking the train often does not get you to the final destination. Unless that destination is in the center city or close to a commuter rail or transit line, a car rental may be required.
C&NW, CA&E, MILW, CGW and IC fan
schlimm The real point underlying these discussions is this question: At what maximum distance and time between city pairs can passenger rail be relevant in terms of market share compared to buses and planes? Time is probably 4-6 hours. An important variable is sustained average speed. If it is 60mph, then 250-350 miles. But if sustained speed is 125, then even 500-650 miles might be competitive. Acela can only sustain 85 mph Wash-NYP. The CZ manages about 55.
The real point underlying these discussions is this question: At what maximum distance and time between city pairs can passenger rail be relevant in terms of market share compared to buses and planes? Time is probably 4-6 hours. An important variable is sustained average speed. If it is 60mph, then 250-350 miles. But if sustained speed is 125, then even 500-650 miles might be competitive. Acela can only sustain 85 mph Wash-NYP. The CZ manages about 55.
I think you mean "average" or "mean" instead of "sustained". But, you are correct - it's largely about trip time - which is a "door to door" measure played off against the cost in hard dollars and soft cost of "hassle".
Here's an ancedote. My college-age daughter and her three friends are taking a trip to Washington DC in a few weeks. They are going to stay with another friend who lives on the outskirts of DC and go to a soccer game and do some sightseeing.
They found some flights from ATL to BWI that rivaled their out-of-pocket cost of driving (their full, true marginal cost is only slightly more - the car they will take has little value). So, fly, right? Not so fast.
On the time side of the ledger, you have the dwell time at ATL, the transfer time to BWI rail station, station dwell, train transit time to Union Station, Metro dwell and ride and finally, a taxi to their friend's house. So, it wasn't a 10 hour drive versus a two hour flight. It was 10 hours door to door versus 7 or 8 hours if they flew.
On the cost side, once they added in the local transit costs, the low air fare wasn't quite so low.
They haven't priced out Amtrak yet. The schedule might actually work for them, but it has some of the problem of local mobility that flying does and the long transit time of driving.
Charlotte has two Piedmont trains a day in each direction connecting it with Raleigh. Also, it is the southern terminus of the Carolinian which runs to and from New York with a connection to Boston. I count that as 6 trains a day.
Charlotte is the 17th largest city in the country with a population of about 751,000. However it has the 6th largest airport because, as you point out, it is a hub for US Air. So while there are many seats on the planes they serve a lot more than Charlotte passengers. Of course the trains -- even the Piedmonts -- serve a lot more than Charlotte passengers too.
The Piedmonts are Amtrak trains but I'm sure North Carolina pays for them.
Many people have talked about Amtrak dying, but it is still alive & surviving. Additional trains have been added between Chicago and St. Louis and when I've been at the station, the trains are packed. Even the River Runner to KC is busy but not all the time.
Ridership is up and that's the bottom line and good news. Many people do not like to fly or don't want the hassle of going through all the commotion at airports today. And St. Louis is no longer a hub, just a spoke, since TWA was destroyed and chopped up. If I can get to where I want to go by train, I will choose that option.
I've been on some of Amtrak's long distance trains and they do a great job of taking care of the passengers. I can compare that with the passenger trains I rode with my parents when they were owned by individual RR's. There are some people in the government that would like to cut Amtrak, Mitt Romney was one of them, but others would never let Amtrak go down, because too many of their voters depend on the service.
John WR oltmanndAmtrak has 0.8% market share of all common carriers". Well yes, Don. But you are comparing Amtrak with all bus and air routes including those where Amtrak has no parallel route and therefore could not possible carry any passengers. If only parallel routes were used the results would be different. Also Amtrak is used as a surrogate for rail transit. But there is rail transit other than Amtrak. For example, it is quite possible to take New Jersey Transit and Septa between New York and Philadelphia and many people do. It is also possible to take Amtrak. Shouldn't such non Amtrak routes be considered? John
oltmanndAmtrak has 0.8% market share of all common carriers".
Well yes, Don. But you are comparing Amtrak with all bus and air routes including those where Amtrak has no parallel route and therefore could not possible carry any passengers. If only parallel routes were used the results would be different.
Also Amtrak is used as a surrogate for rail transit. But there is rail transit other than Amtrak. For example, it is quite possible to take New Jersey Transit and Septa between New York and Philadelphia and many people do. It is also possible to take Amtrak. Shouldn't such non Amtrak routes be considered?
John
Yes. And on the NEC you will see that rail travel is very relevant - even w/o adding in the probable multi-seat rides.
It's everywhere else where things go bad. Just pick one LD train and a major city pair on the route. Say Chicago to Denver. Or, Charlotte to NYC. Then, count the seats available on Amtrak and count the seats available by air. Then divide. It's not really apples to apples since Amtrak serves many more city pairs than those two - and presumably only has a fraction of the seats available for those passengers. Also, Charlotte is a hub city for USAir, so not all their pax are local, either. But, be that as it may....
There are 47 non stop flights from Charlotte to NYC every day versus two Amtrak trains. That's about 15:1 in seats - just for NYC area airports. Amtrak is truly irrelevant in that market at present.
BonasStop Comparing Amtrak ridership with automobiles. You have to use % of "Common Carriers" that charge a fare for a seat like the Airlines and Greyhound.
When I was a student I took a course about creating mathematical models. The professor began by pointing out "The whole truth about a thing is the thing itself." No model can ever be completely accurate. There is no truly accurate way to compare passenger transportation by rail, automobile and airplane. Yet when it comes to public policy we do make decisions about how much to spend on each form and we do need a basis for those decisions. What should the basis of our decisions be?
oltmannd Bonas Stop Comparing Amtrak ridership with automobiles. You have to use % of "Common Carriers" that charge a fare for a seat like the Airlines and Greyhound. Okay. Passenger miles for 2010 (http://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/national_transportation_statistics/html/table_01_40.html) AIr - 564,790 Bus (non transit) - 292,319 Amtrak - 6,470 Amtrak has 0.8% market share of all common carriers". But, to be fair, for 75% of Amtrak's riders, what do you think their #2 choice to "get there" would have been? (remember, 75% of Amtrak riders are on the NEC and other corridors).
Okay.
Passenger miles for 2010 (http://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/national_transportation_statistics/html/table_01_40.html)
AIr - 564,790
Bus (non transit) - 292,319
Amtrak - 6,470
Amtrak has 0.8% market share of all common carriers". But, to be fair, for 75% of Amtrak's riders, what do you think their #2 choice to "get there" would have been? (remember, 75% of Amtrak riders are on the NEC and other corridors).
During FY12 Amtrak's long distance trains carried 15.2 per cent of the system's passengers. Passengers booking sleeping car space accounted for 14.5 per cent of the long distance passengers and 2.2 per cent of system passengers.
Amtrak claims that it has more than 75 per cent of the end point to end point NYC to Washington, D.C. passengers compared to commercial air. I wondered whether the number was truly end point to end point, so I wrote a letter to Mr. Boardman for clarification. His assistant replied that it is indeed true.
Bob,
I appreciate your concern for clarity for guys like me who lack technical expertise. But honestly, Henry writes posts than which none are more clear.
LIRR has there own yard west of NYP and is working on enlarging it.
RIDEWITHMEHENRY is the name for our almost monthly day of riding trains and transit in either the NYCity or Philadelphia areas including all commuter lines, Amtrak, subways, light rail and trolleys, bus and ferries when warranted. No fees, just let us know you want to join the ride and pay your fares. Ask to be on our email list or find us on FB as RIDEWITHMEHENRY (all caps) to get descriptions of each outing.
henry6LIRR is not on these tracks....two tracks from Kearney (Swift) across the Portal Bridge, then at Secaucus Jct. (Lack Int. to Erie Int) there are four tracks for .4. mi through the four platforms to three tracks for .7 then two at from Allied .3 to crossovers at Bergen then duck into the tunnel. Eastbound (based on an older ETT--Nov 2011) 21 trains all arrive at Penn Station West and Bergen on one to three minute headways. Westbound only one of the 18 are DH (X) and are two to six minutes apart at PSW. and Bergen. The lone DH is to Jersey Avenue.
You mistake my point. I'm talking about LIRR coming westbound into Penn Station and stopping at the platforms there. When trains from the west that have just come through the tunnel have discharged passengers and need to move off the platform at NYP, they either have to go westbound... which to me implies back through the tunnel... or eastbound. The question is what's already east of them, or likely to block them, if they go east to a location where they can 'hold' for return after the peak of the rush.
I am quite familiar with the layout of the NYCR, but it's helpful to recapitulate for those who are not.
Overmod henry6 NJT runs 21 eastbound trains into Penn Station, NY between 7AM and 8AM,including there are some Amtrak trains, too. It is a two track railroad but four at Sec. Jct. This addresses some of the 'overlap' involved at the Secaucus stop, but not track capacity 'inbound' through the tunnels. I do not think it unfair to presume that setting up switches to route a given train to its platform can be done in less than CBTC headway, although there may need to be some 'ripple' interlocking action to ensure positive control, so assume that inbound platform space and dwell time are not significant until limits are actually exceeded (probably not at 21 trains/hour inboard... but this ignores all the westbound LIRR stuff going to many of the same platforms at the end needed to 'clear' eastbound consists when they are empty...) LIRR is not on these tracks....two tracks from Kearney (Swift) across the Portal Bridge, then at Secaucus Jct. (Lack Int. to Erie Int) there are four tracks for .4. mi through the four platforms to three tracks for .7 then two at from Allied .3 to crossovers at Bergen then duck into the tunnel. Eastbound (based on an older ETT--Nov 2011) 21 trains all arrive at Penn Station West and Bergen on one to three minute headways. Westbound only one of the 18 are DH (X) and are two to six minutes apart at PSW. and Bergen. The lone DH is to Jersey Avenue. However, there are 18 westbounds during that same hour, so reverse traffic is non existent. I am missing something here. You say there are 21 trains in one hour eastbound, and 18 trains westbound (including deadheads, but not allowing multiple consists to couple and run MU for deadhead moves to give additional headway). And this is on two tracks, over a long enough distance and with limited crossovers so that only a couple of those trains can be safely routed 'counterflow' in that hour. Eastbound (based on an older ETT--Nov 2011) 21 trains all arrive at Penn Station West and Bergen on one to three minute headways. Westbound only one of the 18 are DH (X) and are two to six minutes apart at PSW. and Bergen. The lone DH is to Jersey Avenue. But you just said something about 12 trains per hour being the number for your headway, even assuming short blocks, cab signals, and better speed. It is obvious that even NJT's AVERAGE headway over that hour is considerably less, and I'll bet (without looking at the schedule) that there are a couple of 'knots' where trains must operate on closer spacing to avoid delays. If the average load is 1000 per train, then that is 21,000 people in that hour in one direction. Stipulated. Practically this number per hour could be somewhat greater if the westbounds (especially the deadhead moves or partially-empty counterflow trains) were not prioritized during peak for eastbound rush. That might involve staging trains as far east as possible. One lane of traffic would have to handle up to 14,000 cars (at 1.5 persons per car) per hour at .0042857 seconds apart to handle the equivalent of the railroad. Indeed. But we can't be looking at a one-lane automobile road as the alternative -- that would be like taking the entire flow of inbound auto traffic to Manhattan and routing it through one bore of the Holland Tunnel. I think it is safe to assume that even by 1927 it was understood that multiple lanes would be required even for 'normal' flow density of auto traffic to and from Manhattan during rush hour. So the number of highway lanes has to be increased, How fast is .0042857 leeway? I think you would be talking hundreds of lanes of traffic not just double digits to move at a safe speed For the MCI-9 given, you have 44 nominal seats multiplied by 2000 slots. That's a pretty big number. Relax the assumption to something longer -- say, a three-second following distance, or just under a 4-second headway at 45mph. That gives 900 buses per hour with 44 seats each, which ought to be and is right in there with the density in comparable (single-deck) train service that has much longer necessary headway even with short blocks. Single "deck" or single "track"? It is easier to fill 'every seat' in the train by the time you get to the two-track NYCR 'bottleneck' -- and of course not every bus that runs will have all its seats full. On the other hand, my experience with the buses at the GWB, and the 166 to Manhattan, was that standee conditions right up to max occupancy (and sometimes, ssssh, above it) give much more than nominal seating for the same length and almost the same braking characteristics. So I would not be too smug about the general superiority of rail. An important point that hasn't been mentioned is that the absolute width of railroad 'lane' does not involve any required shoulders or other lateral clearance, since 'steering' is done by default and without particular danger or risk. So there 'ought' to be far less concern over safety than there is with bus lanes ... make the bus guideway physically separate, and the capital concerns need to be factored in... vis-a-vis the existing railroad alternative, but in light of required railroad capacity upgrade costs. Not smug....just have to look at the space available to build new highways...which planners say is none. Then you also have to look at the air quality and fuel consumption. Which points toward trains, especially electric. But also note that Frank J. Martrz Bus of Scranton-Wilkes Barre, PA has had to add almost an hour to his peak hour schedules which is costly for him in wages and fuel as well as rest time for drivers. He likes the idea of train service from Scranton to NY to ease his costs. (Similarly, with the Downeast service out of Portland, ME to Boston, initially the bus company was scared b
henry6 NJT runs 21 eastbound trains into Penn Station, NY between 7AM and 8AM,including there are some Amtrak trains, too. It is a two track railroad but four at Sec. Jct.
NJT runs 21 eastbound trains into Penn Station, NY between 7AM and 8AM,including there are some Amtrak trains, too. It is a two track railroad but four at Sec. Jct.
This addresses some of the 'overlap' involved at the Secaucus stop, but not track capacity 'inbound' through the tunnels. I do not think it unfair to presume that setting up switches to route a given train to its platform can be done in less than CBTC headway, although there may need to be some 'ripple' interlocking action to ensure positive control, so assume that inbound platform space and dwell time are not significant until limits are actually exceeded (probably not at 21 trains/hour inboard... but this ignores all the westbound LIRR stuff going to many of the same platforms at the end needed to 'clear' eastbound consists when they are empty...)
LIRR is not on these tracks....two tracks from Kearney (Swift) across the Portal Bridge, then at Secaucus Jct. (Lack Int. to Erie Int) there are four tracks for .4. mi through the four platforms to three tracks for .7 then two at from Allied .3 to crossovers at Bergen then duck into the tunnel. Eastbound (based on an older ETT--Nov 2011) 21 trains all arrive at Penn Station West and Bergen on one to three minute headways. Westbound only one of the 18 are DH (X) and are two to six minutes apart at PSW. and Bergen. The lone DH is to Jersey Avenue.
However, there are 18 westbounds during that same hour, so reverse traffic is non existent.
I am missing something here. You say there are 21 trains in one hour eastbound, and 18 trains westbound (including deadheads, but not allowing multiple consists to couple and run MU for deadhead moves to give additional headway). And this is on two tracks, over a long enough distance and with limited crossovers so that only a couple of those trains can be safely routed 'counterflow' in that hour.
Eastbound (based on an older ETT--Nov 2011) 21 trains all arrive at Penn Station West and Bergen on one to three minute headways. Westbound only one of the 18 are DH (X) and are two to six minutes apart at PSW. and Bergen. The lone DH is to Jersey Avenue.
But you just said something about 12 trains per hour being the number for your headway, even assuming short blocks, cab signals, and better speed. It is obvious that even NJT's AVERAGE headway over that hour is considerably less, and I'll bet (without looking at the schedule) that there are a couple of 'knots' where trains must operate on closer spacing to avoid delays.
If the average load is 1000 per train, then that is 21,000 people in that hour in one direction.
Stipulated. Practically this number per hour could be somewhat greater if the westbounds (especially the deadhead moves or partially-empty counterflow trains) were not prioritized during peak for eastbound rush. That might involve staging trains as far east as possible.
One lane of traffic would have to handle up to 14,000 cars (at 1.5 persons per car) per hour at .0042857 seconds apart to handle the equivalent of the railroad.
Indeed. But we can't be looking at a one-lane automobile road as the alternative -- that would be like taking the entire flow of inbound auto traffic to Manhattan and routing it through one bore of the Holland Tunnel. I think it is safe to assume that even by 1927 it was understood that multiple lanes would be required even for 'normal' flow density of auto traffic to and from Manhattan during rush hour.
So the number of highway lanes has to be increased, How fast is .0042857 leeway? I think you would be talking hundreds of lanes of traffic not just double digits to move at a safe speed
For the MCI-9 given, you have 44 nominal seats multiplied by 2000 slots. That's a pretty big number. Relax the assumption to something longer -- say, a three-second following distance, or just under a 4-second headway at 45mph. That gives 900 buses per hour with 44 seats each, which ought to be and is right in there with the density in comparable (single-deck) train service that has much longer necessary headway even with short blocks.
Single "deck" or single "track"?
So I would not be too smug about the general superiority of rail. An important point that hasn't been mentioned is that the absolute width of railroad 'lane' does not involve any required shoulders or other lateral clearance, since 'steering' is done by default and without particular danger or risk. So there 'ought' to be far less concern over safety than there is with bus lanes ... make the bus guideway physically separate, and the capital concerns need to be factored in... vis-a-vis the existing railroad alternative, but in light of required railroad capacity upgrade costs.
So those are my remarks...here so that this can be accepted for posting by the machines....
OvermodFor fun: Run the same numbers assuming sesqui-deck Van Hools or whatever (like what MegaBus runs), or articulated sesquidecks that are essentially full bilevel in the forward chassis. Capital cost of those things may be much lower per passenger than dedicated bilevel rail equipment, particularly if the rail equipment has to be built to FRA standards for buff, draft, and empty weight. Also not difficult to provide dedicated electrical power to a BEV-capable bus for the duration of its running in a dedicated bus lane (counterflow or otherwise) -- you'll need to be CAREFUL, but it could be done.
I understand that by "sesqui-deck Van Hools" you refer to one particular brand of double deck buses.
By using only double deck buses it may well be possible to put more people through the tunnels and over the bridges during the rush hour. And if the No. 7 subway is extended to New Jersey additional increases would be possible. Of course the subway is a train but it is not Amtrak, Metro North or New Jersey Transit. But apart from that there is another issue. For years now I've been seeing in the media that the Port Authority Bus Terminal is at capacity and Manhattan's streets also are during the rush hours. There would be no point in adding more vehicles with shorter headways because once they get into Manhattan there is no place more vehicles can go.
OvermodFor fun: Run the same numbers assuming sesqui-deck Van Hools or whatever (like what MegaBus runs), or articulated sesquidecks that are essentially full bilevel in the forward chassis. Capital cost of those things may be much lower per passenger than dedicated bilevel rail equipment, particularly if the rail equipment has to be built to FRA standards for buff, draft, and empty weight.
An new 55 seat bus runs $400,000-$500,000 vs. $2.5-$3.0 M for an FRA compliant rail coach. I don't imagine those VanHools are much different. That's part of our "problem".
OvermodFor the MCI-9 given, you have 44 nominal seats multiplied by 2000 slots.
The newer MCI D4500 commuter buses have 55 seats (w/o a wheelchair rider - they have two wheelchair "slots" that take up six seats). They are quite nice. I commute in one. No seat pitch issues...I have a 35" inseam and can pretty much fit in any seat, even when the guy ahead of me reclines.
henry6 Actually trains can...and in dense traffic areas on passenger railroads actually are...run closer than yhou suggest OLTMANND. In these areas you usually have shorter blocks...maybe a half to 3/4 of a mile, even less on rapid transit lines...plus cab signals and at much higher speeds than 50, up to 79 or more. With speed and stopping distances taken into account you can be running trains at 5 minute intervals which would give you up to 12 trains per hour...
Actually trains can...and in dense traffic areas on passenger railroads actually are...run closer than yhou suggest OLTMANND. In these areas you usually have shorter blocks...maybe a half to 3/4 of a mile, even less on rapid transit lines...plus cab signals and at much higher speeds than 50, up to 79 or more. With speed and stopping distances taken into account you can be running trains at 5 minute intervals which would give you up to 12 trains per hour...
This is the reason I explicitly brought up CBTC, which allows trains to run with MINIMUM safe headway regardless of legacy or default 'block' structure. That will give you more than 12 trains per hour, and I believe some more specific capacity estimates are available specifically for the NYCR north/east of Allied/Secaucus/Lautenberg whatever.
Yes, this makes your point even more dramatically. It's interesting to see what the maximum 'theoretical' capacity of the NYCR is, net of acceleration and deceleration at Allied (we can rule out the pure dwell with the multiple tracks). It should not be surprising that this calculation has been done, and perhaps less surprising that a principal justification for the Gateway project is to open up additional capacity for this period (at the least, by allowing Amtrak service to be separate from NJT locals from all the myriad ways).
Which brings up the peak dedicated-bus-lane numbers again. Use Oltmann's freeway-occupancy stats for dwell, since the additional following distance or stopping time for the buses is balanced by the fact that all the drivers are relatively highly skilled, experienced, and alert/responsive. The assumption (probably justified in this model) that there will be insignificant delay to reach a platform, and relatively large 'parking' capacity for buses that are not needed for deadhead or other purposes (and implicitly that there are enough buses to sustain these conditions during the full time considered).
For fun: Run the same numbers assuming sesqui-deck Van Hools or whatever (like what MegaBus runs), or articulated sesquidecks that are essentially full bilevel in the forward chassis. Capital cost of those things may be much lower per passenger than dedicated bilevel rail equipment, particularly if the rail equipment has to be built to FRA standards for buff, draft, and empty weight. Also not difficult to provide dedicated electrical power to a BEV-capable bus for the duration of its running in a dedicated bus lane (counterflow or otherwise) -- you'll need to be CAREFUL, but it could be done.
henry6Actually trains can...and in dense traffic areas on passenger railroads actually are...run closer than yhou suggest OLTMANND.
I agree. I was referring to freight traffic where you typically have to move an entire mile long train off the main into a yard at restricted speed (typically 15-20 mph). It's a mainline capacity killer.
Actually trains can...and in dense traffic areas on passenger railroads actually are...run closer than yhou suggest OLTMANND. In these areas you usually have shorter blocks...maybe a half to 3/4 of a mile, even less on rapid transit lines...plus cab signals and at much higher speeds than 50, up to 79 or more. With speed and stopping distances taken into account you can be running trains at 5 minute intervals which would give you up to 12 trains per hour....NJT runs 21 eastbound trains into Penn Station, NY between 7AM and 8AM,including there are some Amtrak trains, too. It is a two track railroad but four at Sec. Jct. Normally the two tracks are tr 1 west, tr 2 east but are reverse signaled. However, there are 18 westbounds during that same hour, so reverse traffic is non existant. If the average load is 1000 per train, then that is 21,000 people in that hour in one direction. Fewer people westbound because several of the trains are deadhead moves. That also means that one lane of traffic would have to handle up to 14,000 cars (at 1.5 persons per car) per hour at .0042857 seconds apart to handle the equivelant of the railroad.
Ballpark for autos per hour per lane of freeway is 2000.
OvermodAnd for anyone experienced in freight track-occupancy concerns, where effective over-the-road speed and hence available ton-miles is limited by yard approaches and yard capacities, this will come as no particular surprise.
Yup. You can run'em at 50 mph two blocks apart (~4 miles)...until you have to yard them somewhere. Then, 3 or 4 trains a hour can get tough.
oltmanndWhat absolute nonsense! Particularly the lane of highway.
Nonsense alright*, but not quite in the sense you mean...
The most used highway lane in the country is the inbound Lincoln Tunnel bus-only lane in the morning rush hour, which sees one bus every 45 seconds...
And here I was thinking that you were from New York.
The buses in the bus-only lane, during rush hour, are on MUCH less than "45-second" headways. I do not think it's less than a few seconds, measured in time as opposed to distance -- and yes, it's dangerous as hell, but the drivers manage it nicely.
I do not know the 'official' speed limit in the counterflow lane, but let's call it 30 mph, or 44 feet per second. That's one MCI-9 coach length with a little clearance. There is much less than a bus length visible between buses -- but for grins, call it a whole length. That's a headway of ... two seconds, Split the difference on the seat tracking and call it 45 pax per coach. Or on average, 23 pax per second. That does not produce "4500 passengers per hour".
Put 'radar cruise control' and some intervehicle 'mechatronic' controls to make braking safer, and you can reduce the effective headway somewhat further. It has been 'practical' for many years to run buses essentially nose to tail (with what are essentially spring buffers for absorbing any residual delta due to transient vehicle, road, or weather conditions). Autonomous-control assistance, when perfected, will facilitate this further, but there are diminishing returns to scale in this situation.
With private automobiles, the headway could be reduced to 20 seconds because of faster stopping capability ...
Twenty-second headway? In NEW JERSEY??
Methinks you've been 'upcountry' far too long.
If it's more than about 2 seconds following distance, on that kind of road, I'd be surprised. And at whatever speed would prevail in the absence of blockage (not by any means a safe assumption, and I'll get to it later, but you were talking about unrestricted capacity, not actual Lincoln Tunnel or whatever traffic...) Meanwhile, the average length of a car has gone down fairly dramatically, as have the characteristics requiring longer braking distance to avoid collision, so headways can be effectively shorter for a variety of reasons. True, you're not going to approach bus seating density. But you can easily make that point without having to "lie with statistics."
As a sanity check, the guideway parameters estimated for PRT system 'mains' fall neatly into the middle of this range of occupancy, with a somewhat tighter packaging leading to (slightly) better headway metrics.
, and with 2 people per car, which is a higher than statistical average, we come up with 360 people per hour.
All you have to do is look at that number and you'll see it's ridiculous. I see traffic counts for two-lane roads -- suburban roads -- that are higher than that. By an order of magnitude or more. And that implicitly presumes one passenger per vehicle...
I am assuming, of course, a reasonable speed, 45 miles per hour. If you want to crawl along bumper to bumper, your can get a headway of a car every 10 seconds, but you'll be traveling at 20mph. Even then, only 1200 people per hour.
And that, of course, is the joker in the deck for lane capacity, and the point you SHOULD have emphasized about lane capacity.
On the approach to the Lincoln Tunnel in rush hour, lane capacity is reduced by two notable characteristics: 'average' travel speed somewhere in the 2-5 mph range, and following distance that can vary in length depending on reaction time and acceleration/braking as drivers fail to react to openings. (You can watch this effect clearly on overhead-camera views that have been speeded up). So you have variable headway even (particularly, in most circumstances!) at slow speeds or in aptly-named 'stop and go traffic'.
As Frederik Pohl noted, "a pipeline has two ends". And if the inner end is blocked or impeded, and the fluid in it is incompressible beyond a certain point and statistically resistant to compression for a ways above that, it does not take a math or physics degree to understand that flow, by any measure, is implicitly reduced. On roadways this occurs almost always in conjunction with periods of peak demand (which is when any measure of occupancy has particular meaning in a comparison analysis).
The traditional methods of highway capacity expansion (multiple lanes, dedicated flow lanes alternate routes) have no particular applicability to a lane-capacity analysis in this context.
Now, meanwhile, over on the NYCR, the situation is not static. With the assumption of CBTC, effective headway shrinks to 'electronic' following distance, even at higher track speed. "Throughput" in this situation is time-limited in a different sense; due to (1) required reduction in speed for the east end of the tunnels, negotiation of the yard to arrival tracks, and (2) dwell time on available tracks, plus time to clear them after deboarding. This was as I recall a big piece of why THE Tunnel was so complex and expensive on the New York end. And for anyone experienced in freight track-occupancy concerns, where effective over-the-road speed and hence available ton-miles is limited by yard approaches and yard capacities, this will come as no particular surprise. The 'busway' proposal has a theoretical advantage: at the receiving end there is a multiplicity of separate routes going to separate platforms, all of which essentially operate in parallel -- there is no trudging down a long platform in a sea of people. (This in part makes up for the difficulty posed by over-the-road buses usually lacking rear or center doors, but theoretically there is no engineering reason I know of why low-floor artics would not be suited for counterflow bus lanes with high-occupancy control systems, and while admittedly there are some placement difficulties over doors in the rear section of artics, you could easily put in a short section of high-level platform, or a lift, to access the area over the engine and drive...)
None of this is intended in any way to subtract from the original point, which is that one 'lane' of unidirectional rail can carry more traffic easily than any rubber-tired route short of something like a Metro. Especially when overhead clearances permit full bilevel cars to run. I had thought that just an inspection of the 'packing fraction' of the plan would tell people that. It's just not as bad as the provided numbers would indicate.
* this is the one rhetorically permissible use of 'alright' instead of 'all right'... if it galls you think of it as a colloquialism.
daveklepper What absolute nonsense! Particularly the lane of highway. The most used highway lane in the country is the inbound Lincoln Tunnel bus-only lane in the morning rush hour, which sees one bus every 45 seconds, and is estimated to have 4500 passengers per hour, some standing. With private automobiles, the headway could be reduced to 20 seconds because of faster stopping capability, and with 2 people per car, which is a higher than statistical average, we come up with 360 people per hour. I am assuming, of course, a reasonable speed, 45 miles per hour. If you want to crawl along bumper to bumper, your can get a headway of a car every 10 seconds, but you'll be traveling at 20mph. Even then, only 1200 people per hour. The PRR tubes under the Hudson River handle a train every two mintues with an average occupancy of 1000 people per train, for 30,000 passengers. Admittadly these are commuter trains with standees, mostly. If they were all long-distance trains, the passenger count would drop to 10,000,, still a lot more than any highway lane. You ought to compare yourself with a truck driver who says trucks should replace railroad freight because trucks are more efficient and can go anywhere and don't require loading on flatcars or transfering frieght from a truck to a train and back to a truck. His coment would be as stupid as yours but no stupider. Sure comparing the most efficient auto with the average of Amtrak makes the auto come out ahead. But comparing the average auto with Amtrak's average shows Amtrak ahead, and comparing the most efficient Amtrak operation with the most efficient auto also shows Amtrak ahead.
What absolute nonsense! Particularly the lane of highway. The most used highway lane in the country is the inbound Lincoln Tunnel bus-only lane in the morning rush hour, which sees one bus every 45 seconds, and is estimated to have 4500 passengers per hour, some standing. With private automobiles, the headway could be reduced to 20 seconds because of faster stopping capability, and with 2 people per car, which is a higher than statistical average, we come up with 360 people per hour. I am assuming, of course, a reasonable speed, 45 miles per hour. If you want to crawl along bumper to bumper, your can get a headway of a car every 10 seconds, but you'll be traveling at 20mph. Even then, only 1200 people per hour.
The PRR tubes under the Hudson River handle a train every two mintues with an average occupancy of 1000 people per train, for 30,000 passengers. Admittadly these are commuter trains with standees, mostly. If they were all long-distance trains, the passenger count would drop to 10,000,, still a lot more than any highway lane.
You ought to compare yourself with a truck driver who says trucks should replace railroad freight because trucks are more efficient and can go anywhere and don't require loading on flatcars or transfering frieght from a truck to a train and back to a truck. His coment would be as stupid as yours but no stupider.
Sure comparing the most efficient auto with the average of Amtrak makes the auto come out ahead. But comparing the average auto with Amtrak's average shows Amtrak ahead, and comparing the most efficient Amtrak operation with the most efficient auto also shows Amtrak ahead.
It's interesting that most of us (with one notable exception!) who snipe at each other about this and that pretty much come down on the same side vis a vis the NEC.
ontheBNSF oltmannd ontheBNSFThe NEC project is more expensive than the Japan Maglev project. 6 Billion for a station just for a station come on. Uh, "just a station?" That's pretty funny! But seriously, what's the alternative? The alternative is a less fancy station which just adds extra capacity.
oltmannd ontheBNSFThe NEC project is more expensive than the Japan Maglev project. 6 Billion for a station just for a station come on. Uh, "just a station?" That's pretty funny! But seriously, what's the alternative?
ontheBNSFThe NEC project is more expensive than the Japan Maglev project. 6 Billion for a station just for a station come on.
Uh, "just a station?" That's pretty funny! But seriously, what's the alternative?
The alternative is a less fancy station which just adds extra capacity.
Exactly how much of the $6B is for "fancy"?
daveklepper What absolute nonsense! [in reference to the original poster's contention]
What absolute nonsense! [in reference to the original poster's contention]
Well said!!
Our community is FREE to join. To participate you must either login or register for an account.