neroden
Engineer
Yeah...I wonder if they are constantly sold out, they added ANOTHER sleeper on, AND they added the diner back as cash only what the results would be.
So, I just ran a very rough calculation. Assume that by losing the diner, Amtrak is losing $120 per roomette and $240 per bedroom per night (roomette number from other people's measurements, bedroom number a WAG based on the roomette number). But assume that Amtrak is gaining an average of $200 per roomette per night by selling the rooms formerly used by the crew and freeing up 5 rooms (WAG number based on intermediate bucket). This would be something like $1.7 million a year lost in revenue, net. (Multiply daily amounts by 365 x 2 for north and south departures. This isn't including any lost coach revenue, but so far there isn't much evidence of significant lost coach revenue on the Star.)
But my latest attempt to compute dining car expenses comes out to about $2.2 million / year (before accounting for lost revenue from occupied roomettes, since I counted that above); $100K per staff member x 5 staff/train x 4 trainsets, $50K maintenance per car x 4 trainsets. (I'm ignoring both the food -- as negligible -- and the coach payments for the food.)
(PS -- thanks Paulus for giving me some sort of order of magnitude number to use for car maintenance on more modern cars. Heritage car maintenance may be significantly higher than that, but probably not more than double, which would come to 2.4 million.)
Now redo the calculations with *three* sleepers. The lost revenue from losing the dining car (net) becomes $3.3 million. The dining car expenses remain $2.2 million. Suddenly the dining car is financially beneficial.
Economies of scale!
Perhaps two sleepers and one diner is worse financially for Amtrak than two sleepers and no diner.
But three sleepers and one diner is better financially for Amtrak than three sleepers and no diner.
And I think we're in consensus that on routes where the demand is there, three sleepers is better financially than two sleepers...
Looks to me like the Star should be beefed up to three sleepers along with the restoration of its dining car. If the Cardinal ever goes daily, as it should, it is likely to develop enough demand to be beefed up to three sleepers + a dining car too. (The Crescent may benefit from this as well perhaps, though it has the problem of weak demand south of Atlanta.)
Conclusion: All single-level trains with dining cars should have three sleepers minimum. Amtrak can't get the Viewliner IIs soon enough!
----
And frankly Amtrak needs even more sleepers. ASAP. Adding a 4th sleeper to the Meteor, one sleeper (no diner) to #66/67, one through sleeper on the Pennsy to the Capitol Limited, 3 sleepers to Atlanta and 1 to New Orleans on the Crescent, 3 sleepers daily on the Cardinal, 3 sleepers on the Star, 4th on the LSL or 2nd on the Pennsy, add 20% shop time and even with the Viewliner IIs, there are no cars left for protect cars.
A reasonable proposal of 3 daily sleepers on every existing eastern train with sleepers, 4 on the Meteor, 1 on #66/67, 3 on the Pennsy/Cap, 6 protect cars, and 20% shop time, gives 90 sleepers desired as opposed to 75 (existing and ordered).
To convert the Cap to single-levels (desirable for consistency with the Pennsy and consistency on the Ohio/Indiana platforms) would call for another 12 sleepers or so; this would also take some pressure off the Superliner sleeper, diner, and lounge fleets. I think there are enough coaches to handle a single-level Cap for now, particularly as the Horizons come free and can be refurbished.
So Amtrak has 50 sleepers and is ordering 25 more; Amtrak needs another 27 on top of that. The original proposed order of 100 back when the Viewliner Is were ordered was about right.
(I also come up with immediately useful applications for 5 more dining cars, based on the above.)
20% shop time may be too much, or maybe protect cars don't need shop time, which would cut these numbers some. Use of bag-dorms might cut the number of sleepers needed a little as well. But it still doesn't change the result that more sleepers are needed.
Last edited by a moderator: