Amtrak selling intra NEC seats on SB LD trains

[the_traveler]

I wonder how exactly this will work though. The simple solution is to just open up a number of seats and sell what sells. The more advanced solution, is to sell the seat, only after a seat from that station to off the corridor has been sold. So a seat BAL-ATL, opens up a seat NYP-BAL, or PHL-BAL which in the latter case still gives a NYP-PHL seat. I don't know if Arrow is capable of that or not though.

I don't see why Arrow can't do that. It can sell CHI-KCY, KCY-ABQ, ABQ-FLG and FLG-LAX on the SWC.

 

[johnny.menhennet]

I entirely agree with MattW ont his matter. What I have ALWAYS thought would work best is that on southbound trains, for every ticket that is sold Alexandria-point south, open up a slot NYP-ALX, or with WAS as well, at first opening of seats, dont sell any tix for intra-NEC traffic, but when a ticket WAS-MIA is booked, allow one seat to be booked from WAS to points north. For the poster concerned about dining space being taken up by these NEC riders, i know that at least with the CS southbound, they take all dinner reservations right before arrival into SBA, and SBA pax are forced to eat from the lounge or from a local place before the train (always our option). I realize this would work only on a northbound run, really, and I'm not sure how I feel about northbound trains losing their ability to arrive early, but... i can't think of a great solution to that. southbound the opening of a seat once an LD one is purchased is completely feasible.

 

[benjibear]

Maybe these seats on the LD trains should be held in reserve. Only open them up once the prices on the NER is at a higher bucket and charge the higher bucket on the LD trains. This may be timing. The day a train goes up for sale, you can't purchase WAS-NYP but closer to departure it opens up. This would keep the LD trains primarily LD but at busy times, would allow Amtrak more capacity.

I am not sure Amtrak would be arriving early into DC northbound. I see the opposite happening and having unhappy passengers.

 

[amamba]

Only thru sept 30, if you want them...
NYP-WAS on the star is $145

NYP-MIA on the star is $141 same date.

Wow that price is ridiculous. I booked PVD - ALX on a regional connecting to the silver star in NYP - WITH a roomette - for $156. I am going in August.

 

[afigg]

Another thing I just thought about, that ought to compel Amtrak to sell space on NB LD trains is the timing. Unlike their SB counterparts, where a train must be held at the station if early because things are working better than normal for that day, the Northbounds often make timings that are not far off the Regionals, owing to their not stopping at so many stations.

Today, Thursday, July 18, is a good example of the problem with selling seats on the northbound LD trains for WAS to NYP.
#20 (7/17) Crescent: departed WAS 2 hours 35 minutes late (arr NYP 3:13 late!)

#92 (7/17) Silver Star: departed WAS 4 hours 35 minutes late

#80 (7/18) Carolinian: departed WAS 1 hour 4 minutes late

#90 (7/18) Palmetto: departed WAS 57 minutes late

if the Meteor is included in the intra-NEC sales

#98 (7/17) Silver Meteor: departed WAS 2 hours 28 minutes late

Would be a lot of unhappy NEC customers if they brought tickets in advance and there were not enough seats on other Regionals to put them on at the last minute.

In a few years, perhaps the Carolinian will be become more reliable once the track projects on the NC Piedmont corridor are completed, if the 2 HSIPR projects in VA and NC for the CSX tracks get out of limbo and are done, VA pays for track upgrades between ALX and Petersburg. Then #80 could be open for WAS to NYP sales. But the northbound LD trains? Not likely.

 

[gmushial]

I wonder how exactly this will work though. The simple solution is to just open up a number of seats and sell what sells. The more advanced solution, is to sell the seat, only after a seat from that station to off the corridor has been sold. So a seat BAL-ATL, opens up a seat NYP-BAL, or PHL-BAL which in the latter case still gives a NYP-PHL seat. I don't know if Arrow is capable of that or not though.

From a software perspective, it wouldn't be difficult... the question would be if management was willing / had the foresight to do such.

 

[jebr]

I wonder how exactly this will work though. The simple solution is to just open up a number of seats and sell what sells. The more advanced solution, is to sell the seat, only after a seat from that station to off the corridor has been sold. So a seat BAL-ATL, opens up a seat NYP-BAL, or PHL-BAL which in the latter case still gives a NYP-PHL seat. I don't know if Arrow is capable of that or not though.

From a software perspective, it wouldn't be difficult... the question would be if management was willing / had the foresight to do such.

Or if implementing it into ARROW is both A) worth the time for testing, debugging, and making sure it works (if it's even possible on that piece of software) and B) won't cause issues elsewhere in the software.

At least based on what I've read here, ARROW is some pretty old software that's basically just being used as-is, even though they would like to have a better ticketing system in place. (For example, why does 7/27 and 8/28 need to be displayed as separate trains from SPK-CHI? Apparently because ARROW can't work any other way.) Could be way off, though.

 

[gmushial]

I wonder how exactly this will work though. The simple solution is to just open up a number of seats and sell what sells. The more advanced solution, is to sell the seat, only after a seat from that station to off the corridor has been sold. So a seat BAL-ATL, opens up a seat NYP-BAL, or PHL-BAL which in the latter case still gives a NYP-PHL seat. I don't know if Arrow is capable of that or not though.

From a software perspective, it wouldn't be difficult... the question would be if management was willing / had the foresight to do such.

Or if implementing it into ARROW is both A) worth the time for testing, debugging, and making sure it works (if it's even possible on that piece of software) and B) won't cause issues elsewhere in the software.

At least based on what I've read here, ARROW is some pretty old software that's basically just being used as-is, even though they would like to have a better ticketing system in place. (For example, why does 7/27 and 8/28 need to be displayed as separate trains from SPK-CHI? Apparently because ARROW can't work any other way.) Could be way off, though.

Legacy code comes in all flavors - some was a hack from the day it was written and has been a problem every day since... conversely I can think of some software from the late 70s which has been exceedingly useful, bug-free and is basically outliving the hardware it was designed to run on. Somebody with knowledge of the structure and internals of ARROW would be able to answer better.... but at least from a theoretical perspective, what you're asking is not difficult (assume the existing code isn't fighting you at every turn).

 

[Swadian Hardcore]

I'm glad to see Amtrak doing this. Means more revenue and ridership for the LD trains, even on a SD sector. The fares deserve to be higher because these trains are more comfortable.

 

[calwatch]

To discourage northbound traffic (much like they are discouraging southbound traffic on long distance trains), simply place all of those seats in the high bucket. If you want to pay twice the prevailing price, no one will stop you. Also by charging the highest rate possible, you can allow for free changes to Regional trains. To allow for free running time maybe you have the discharge only period start from Baltimore, instead of Washington. Baltimore-NYP is still a good sized trip.

 

[railiner]

I wonder how exactly this will work though. The simple solution is to just open up a number of seats and sell what sells. The more advanced solution, is to sell the seat, only after a seat from that station to off the corridor has been sold. So a seat BAL-ATL, opens up a seat NYP-BAL, or PHL-BAL which in the latter case still gives a NYP-PHL seat. I don't know if Arrow is capable of that or not though.

From a software perspective, it wouldn't be difficult... the question would be if management was willing / had the foresight to do such.

Or if implementing it into ARROW is both A) worth the time for testing, debugging, and making sure it works (if it's even possible on that piece of software) and B) won't cause issues elsewhere in the software.

At least based on what I've read here, ARROW is some pretty old software that's basically just being used as-is, even though they would like to have a better ticketing system in place. (For example, why does 7/27 and 8/28 need to be displayed as separate trains from SPK-CHI? Apparently because ARROW can't work any other way.) Could be way off, though.

The way Arrow works, an agent can work around 'sold out' thru space, by combining segments from the two or more separate trains, and then manually pricing it at the thru fare. At least that's how it used to work. Back when trains 5, 25, and 35 were combined from CHI to SLC, it helped several get a booking, that was 'sold out'...

Not sure if this can still be done....

 

[gmushial]

I wonder how exactly this will work though. The simple solution is to just open up a number of seats and sell what sells. The more advanced solution, is to sell the seat, only after a seat from that station to off the corridor has been sold. So a seat BAL-ATL, opens up a seat NYP-BAL, or PHL-BAL which in the latter case still gives a NYP-PHL seat. I don't know if Arrow is capable of that or not though.

From a software perspective, it wouldn't be difficult... the question would be if management was willing / had the foresight to do such.

Or if implementing it into ARROW is both A) worth the time for testing, debugging, and making sure it works (if it's even possible on that piece of software) and B) won't cause issues elsewhere in the software.

At least based on what I've read here, ARROW is some pretty old software that's basically just being used as-is, even though they would like to have a better ticketing system in place. (For example, why does 7/27 and 8/28 need to be displayed as separate trains from SPK-CHI? Apparently because ARROW can't work any other way.) Could be way off, though.

The way Arrow works, an agent can work around 'sold out' thru space, by combining segments from the two or more separate trains, and then manually pricing it at the thru fare. At least that's how it used to work. Back when trains 5, 25, and 35 were combined from CHI to SLC, it helped several get a booking, that was 'sold out'...

Not sure if this can still be done....

Given how few trains Amtrak actually runs (from a computational perspective) and that we're not talking about some monster combinatorics, it sounds like, again: such wouldn't be terribly difficult to implement (again, assuming the rest of the software isn't malicious)... but again, with the Congress-thingies wishing to strangle Amtrak for their own self-serving political purposes - the question becomes, does Amtrak have even that little money to throw at doing such?

 

[cirdan]

NB LD trains are free to run early on the NEC (and sometimes do). Can't see how it's possible to handle intra-NEC passengers unless Amtrak abandons the ability to run early.

It depends how valuable that option is.

If raking in extra cash outweighs the possible disadvantage of not arriving before the scheduled time, maybe its a fair deal.

 

[Anderson]

On space, Amtrak should "know" they're going to get a certain amount of traffic added to the LD trains at WAS/ALX heading south net of any discharges. You could probably pour through the data for the last year (or five, or twenty, or forty-odd) and find ranges that are either never exceeded or that are exceeded very rarely and under unusual circumstances (i.e. the volunteer crossing guard train that Palm Beach basically buys out once per year). So you can model approximations of those limits, open those seats, and then adjust if sales patterns go in a different direction.

On the issue with late trains: I'd say that at the very least, those trains should be opened at super-peak travel times (i.e. Christmas and Thanksgiving) with stated caveats ("this is an LD train, it may run late and does not have wifi at present") and restrictions on guaranteed connections. Amtrak already runs those "holiday extras" with NJT equipment, and I think there are more than a few folks that would trade being an hour late for not being stuck in a commuter car from DC to New York.

 

[cirdan]

On space, Amtrak should "know" they're going to get a certain amount of traffic added to the LD trains at WAS/ALX heading south net of any discharges. You could probably pour through the data for the last year (or five, or twenty, or forty-odd) and find ranges that are either never exceeded or that are exceeded very rarely and under unusual circumstances (i.e. the volunteer crossing guard train that Palm Beach basically buys out once per year). So you can model approximations of those limits, open those seats, and then adjust if sales patterns go in a different direction.
On the issue with late trains: I'd say that at the very least, those trains should be opened at super-peak travel times (i.e. Christmas and Thanksgiving) with stated caveats ("this is an LD train, it may run late and does not have wifi at present") and restrictions on guaranteed connections. Amtrak already runs those "holiday extras" with NJT equipment, and I think there are more than a few folks that would trade being an hour late for not being stuck in a commuter car from DC to New York.

In theory yes.

But how many people don't read the small print.

And image the negative exposure a severely delayed train would make on a first time rider

Especially if the train is not only late, but after the long trip, stocks on the dining car are depleted, trashcans are full, things are no longer as clean as they could be etc.

 

[Bus Nut]

Likely a (very smart) pilot program on Amtrak's part. I'd raised this idea on here some time ago, but...
However, this move also smacks of shenanigans: Doing this will likely allow Amtrak to pump up ridership on the Star, Crescent, and Cardinal, all of which are off noticeably for FY2013 as far as May (5.4%, 4.6%, and 3.6% respectively). Note that the Meteor, which is basically "breaking even" YTD, is apparently not being sold. If this is successful, it's a "quick fix" for LD ridership for FY13 (presuming the ridership is credited to the LD trains and not to the Regionals). Also note that FY13 ends on...wait for it...Sept. 30.

I'm totally guessing here but I think it's less shenanigans than a really smart way to get the crews/union on board with and enthused about adding NYP-WAS passengers to the SBs. If this actually works to boost ridership and revenue and creates a shiny gold star for these routes this year, then the crews are going to be more excited about such patrons and less annoyed at them for interrupting their zen boarding and settling in the LD patrons. I can imagine that for both the engineer/conductors and the on-board services staff that the quick turnaround patrons on top of the typical demanding LD patrons probably seems like more trouble than it's worth. But if it starts to put them ahead performance-wise within the organization they will see the value.

The other benefit of doing it this way is that if it turns into a disaster (and I'm sure plenty of their own people are predicting doom and gloom) then Amtrak can quickly pull out and tell everybody "See? No sense in doing this, doesn't work."

 

[Nathanael]

At least based on what I've read here, ARROW is some pretty old software that's basically just being used as-is, even though they would like to have a better ticketing system in place.

Yes. Amtrak still has a line item in their budget for replacing ARROW. It would make sense for the replacement software to allow for such things as we have described.

 

[Bus Nut]

To discourage northbound traffic (much like they are discouraging southbound traffic on long distance trains), simply place all of those seats in the high bucket. If you want to pay twice the prevailing price, no one will stop you. Also by charging the highest rate possible, you can allow for free changes to Regional trains. To allow for free running time maybe you have the discharge only period start from Baltimore, instead of Washington. Baltimore-NYP is still a good sized trip.

No no no no no...

1. you're charging more for an unreliable train. But NE corridor riders value on time departure over seat comfort every time.

2. you're charging more and offering an overcrowded corridor train as a backup? Not only is this inconvenient, cannibalizes Acela, and ticks off your patrons twice (first because their scheduled train is late, second because they're jammed onto an overcrowded, slow, less comfortable and probably less clear Regional? What do you give them to make it up? I'm going to assume free travel on top of a refund (not going to repeat what the NYC vernacular would suggest at this point, as it would be NSFW)

3. Now I'm on a Regional, I paid twice what the guy next to me paid. Do I get to kick him out of his seat? But he bought his reserved ticket 3mo.s before I did and thinks I'm a jerk too. We both think the other is a big jerk and say so. Which conductor gets to separate us when we start a shoving match over the seat? :giggle:

4. Washington is a big crazy station with baggage activity and everything else. By comparison, picking up a few punters from BAL-NYP is probably not such a big deal operationally when you're dropping at those stops anyway. So you're kind of saving nothing by embargoing those stops but allowing WAS.

 

[jis]

The other obvious alternative is to offer no advance reservations on the LD trains, but open them up say 24 hours before departure, or even just on the morning of the departure, so walk ups can get on them, and those that haven't found another way can use as a last resort get on them. At that point Amtrak should have absolute firm idea about how many seats they can release. Also if the offered fare is high bucket anyway, there would be no incentive for people to game the system. It will genuinely help those who at the last moment want to travel.

People who just do LD trains have no idea how often people just walk upto a QuikTrak machine in a station and just buy a ticket (+reservation of course) and board a train leaving in the next hour or two, notwithstanding the higher bracket fares that come with it.

Yesterday on the Acela I was on there was a group of some 15 Merrill Lynch folks who said they just walked into Union Station at Washington and bought tickets then and there and boarded the Acela to get back home. 8 of them got off at Metropark! I asked one of them why not reserve earlier? She said, they had absolutely no idea when their workshop would actually end, so they figured since they'd probably have to deal with ticket changing anyway, why bother with the early booking.

So while we endlessly discuss how Amtrak is losing riders to Megabus, one needs to take into the walk up and ride clientele that exists on the NEC. Frankly I was surprised. But then again, even people like me and Piotr have walked into Penn Station, noticed a random reasonable fare available on the evening Acela or Regional to Metropark and just bought a ticket and got on. I have done the same from New Haven too. So you don't necessarily have to be super rich to do that.

Bottom line is, I think even if the LD seats were released very late, there will be significant usage of them. Walk ups simply pull up the list of trains available in the next hour or two and select from them. They really never look at the schedule. As long as QuikTrak is able to bring up trains at a reasonably dependable time of arrival a majority of times people will not be upset. Afterall on days like yesterday at Wilmington absolutely nothing was less than half hour late (Acelas) and Regionals were usually 45 mins to one hour late.

 

[Ryan]

Sounds like a good plan.

The only hitch I see is that there probably isn't a tie in that takes train status into account.

I'd be somewhat annoyed had I walked up at WAS, seen "Hey, I can leave on the Silver Star in 30 minutes for a reasonable price!", booked said ticket and then walked into the waiting area to see that the train is actually running 5 hours late.

 

[jis]

Sounds like a good plan.
The only hitch I see is that there probably isn't a tie in that takes train status into account.

I'd be somewhat annoyed had I walked up at WAS, seen "Hey, I can leave on the Silver Star in 30 minutes for a reasonable price!", booked said ticket and then walked into the waiting area to see that the train is actually running 5 hours late.

Exactly! That is the part I am not sure about. Too bad if that cannot be arranged. Without that, this idea will not work too well. Oh well.

 

[Nathanael]

One would hope that the replacement for ARROW would allow for allowing that kind of walk-up ticket (i.e. start allowing ticket sales automatically only when the train has reached Alexandria, or something). I'm pretty sure the current system would make it very difficult to do that. Not impossible, but requiring too much manual intervention to be worth it except on the occasional holiday or emergency.

 

[me_little_me]

Especially if ... things are no longer as clean as they could be etc.

What do you mean "if"? You mean "since in coach by the time the train reaches Washington they always are" instead of "if".

 

[xyzzy]

The irony is, ALX was marked detrain-only for 20, 92, and 98 not that many years ago. Ever since, those three trains can run way early or way late on the NEC -- particularly 92 and 98, whose last fixed time applies at RVR. If these three trains are on-time at an NEC station, it's just a coincidence. Therefore it's impractical for intra-NEC riders to depend on them. Amtrak apparently prefers to have the operational flexibility to move these three trains along however the dispatchers find it convenient to.

80 and 90, however, do accept pax at ALX. But the Carolinian and the Palmetto are really just a southward extension of the NEC.

Looking south, all trains (19, 79, 89, 91, and 97) are open to pax who detrain at ALX. If one's destination is in metro DC, it's already an alternative to ride an LD train to ALX and catch WMATA from there. But I suspect Amtrak manipulates buckets so that SB shorts to ALX don't block longer sales.

 

[Acela150]

Looking at the Ticketing. This is clearly a Pilot Program. October 1st The Palmetto, SS, and Crescent are not for booking. The Carolinian is, but it has always been open for booking between NYP and WAS.