If it costs more in fuel and labor, that’s almost instantly a reason not to do it, especially given your comment “if Amtrak wants to cut costs, this seems like an easy way to do it.” Increasing costs is not really a good way to cut costs.
Fleet interchangeability isn’t really that big of a problem, given that the fleets tend to stay in their same regions.
Amtrak seems to be moving towards a more simplified fleet with the help of state partners, as in a few years the NEC, Midwest, Cascades, and San Joaquins will all be using some variation of the Siemens Venture cars. Even then, however, you won’t be able to take a car from California and run it from Portland to Seattle.
...................
A properly funded and equipped Amtrak should have no problem efficiently maintaining a bi-level and a single-level fleet, just as they have no problem maintaining separate fleets of diesel and electric locomotives. The problem comes when you have a limited number of unique equipment sets that can only be used in a certain context (i.e. Talgo), where you need to have spares on hand, but simply doing so brings your spare ratio to a ridiculously high level. Once you have a “big enough” fleet to support its own mechanical and operational ecosystem, it really doesn’t matter that much from day to day if that is different from another fleet somewhere else.
True, you can’t run a Surfliner car on the NEC, but why would you ever need to?
For readers who haven't had to schedule mixed fleets that point about spare ratios is important. In the early days of wheelchair-accessible transit buses my predecessor at RTD did a study to determine what was needed to be
almost certain that a trip assigned a lift-equipped bus would turn up. The management had to back off of the enthusiastic initial assignments.
There are some rules of thumb, but you really have to look closely at each need. B.F.
Biaggini told me that an RDC was a poor choice for the SP's stub-station commute service because their single unit was sitting in the shop all the time. They had to substitute a locomotive and a coach. I didn't have the nerve then to point out to him that the CP, CN and B&M all seemed to be able to operate commute service with a big enough fleet.
The SP&S needed one sleeper-lounge to cover PDX<>SPK. They bought two initially when the schedule required it and when the schedule only required one they disposed of a heavy-weight sleeper-lounge that they had kept around just in case.
In the oil boom of 1970's, Edmonton Transit got down to a 9% spare ratio with its obsolete Canadian Car trolley coaches and GM Diesels with no lifts, no registering fareboxes, no radios, no air-conditioning, etc. It was very stressful, but we did it.
My McGraw-Hill electric railway handbook from the 1920's recommends a 10% spare ratio but mentions that some big cities needed 12% spare ratios due to fender-bender accidents.
Today's transit industry standard, with all of the systems loaded and working, is 20%. But if a community wants a cute route with a specially painted bus, it's either buy three buses to cover two runs or else suffer the pain of a generic paint scheme turning up from time to time.
For Amtrak then, a subfleet needs to be large enough to be backed up by say 20% spares. And if it wants to avoid Talgo, it needs an additional 5% or so that can be available for state-sponsored service or special trains, instead of telling funding legislators that they must wait while cars for an experiment are built.
Without the right percentage of spares and the right number of assignments for each subfleet, Amtrak will keep stealing cars to cover work they weren't designed for or will blow off revenue opportunities. And the
Texas Eagle will continue to be victimized.
