Tuesday, August 21, 2012

Delving further into Amtrak route accounting

Some of the strangeness I found by applying train-miles to Amtrak's financial data in my previous post can be explained with one simple answer: Amtrak considers most of their trans that run in the Northeast Corridor as Northeast Regionals even if they're actually other services. However, it doesn't provide a full explanation.

There are 11 Amtrak routes which run significant distances in the Northeast Corridor: the Acela Express, and Northeast Regional are the core services, of course, but there are also the corridor runs for the Vermonter, Keystone Service, Pennsylvanian, and Carolinian, in addition to the long-distance Silver Star, Cardinal, Silver Meteor, Palmetto, and Crescent.

In my previous post, the long-distance trains already seemed to be either on-target or getting overcharged for the number of seats on the route, so they don't appear to be getting reallocated as Northeast Regionals. However, some routes look significantly closer to reality when NEC train mileage is deducted. In particular, the Keystone Service trains to Harrisburg, Pennsylvania and the Pennsylvanian to Pittsburgh get down to variances of 15 to 21 seats—less than half a carload—so we seem to be able to dismiss those two.

The Vermonter, which was showing only 87 seats through my math but runs 5-car consists, is also helped significantly because 37% of its route is in the NEC. Unfortunately, 87 seats was 74% below my capacity estimate, so only a portion of the deficit was accounted for by subtracting NEC route-miles. The Vermonter also shares about 10% of its total route with a shuttle service between New Haven, Connecticut and Springfield, Vermont which might be used to share some accounting, but subtracting those route-miles as well brings me up to a calculated 166 seats per train. Even a train running 4 long-distance coaches and a completely dedicated cafe/lounge car would be expected to have at least 236 revenue seats.

The Carolinian rides on the cusp of being believable now, about 77 seats away from a previous guess I'd had. A little more or a little less than one car load, depending on the type of equipment used.

But four of the routes I mentioned in my earlier post don't really touch the Northeast Corridor at all. The Ethan Allen Express, Maple Leaf, Empire Service, and Adirondack do share some track in the Empire Corridor, and it appears that some similar accounting methods might be getting used to shift train mileage around in those corridors. Subtracting 42% of the Adirondack's mileage and 67% of the Ethan Allen Express route and adding them to the Empire Service's values seems to at least get the first two services in line, but still doesn't explain the Maple Leaf's apparent overcounting of seats: If I shift train-miles around in the same way, I end up with an outlandish 3,591 seats per train rather than the already impossible 554.

Sunday, August 19, 2012

Weirdness in Amtrak seat-miles

I've been busy staring at YouTube videos of trains lately, not because they're all that exciting, but because I'm trying to figure out what typical consists are for various Amtrak routes. A few videos have been fun, though, like this one featuring Metro-North Railroad maintenance-of-way crew moving through after the fourth train. It keeps going, and going, and going...

Anyway, thanks to Paul Druce's table of annual train-miles from earlier this year, it's possible to calculate average passenger and seat counts with some level of precision.

The good news is that this seems to solve one question Paul had about why different services seem to have such a variation in the cost charged per unit distance: There's a strong correlation between numbers of seats and cost per train mile.

The bad news is that the number of seats per train seems really bogus in a lot of circumstances. I've been fumbling my way around Wikipedia, some railroading forums, and spending time watching a lot of videos to try and determine typical consist sizes, and therefore actual seat counts. It's an error-prone process, but I think I've put together some figures which are at least in the ballpark.

The numbers I've calculated for what Amtrak is recording for seats-per-train are also subject to some error. I know there is some level of error in my calculations because I'm working backward from figures for overall contribution/loss per route and contribution/loss per passenger-mile found in Amtrak's September 2011 year-end report (FY2011), but most of those figures have at least 2 digits of precision, and would only be off by a small amount. There are a few situations where Amtrak's figures are only known to 1 digit of precision, but that should still generally be less than a ±10% variance from reality (or reality according to Amtrak, at least).

Similarly, the train-mile calculations I'm using are only accurate to within a few percent because of the method of calculation (number of trains per week multiplied by 52 weeks, though that gives some allowance for canceled trains).

Still, given all of those problems, several of Amtrak's corridor trains in the Northeast (though not necessarily NEC trains themselves) stand out as having highly questionable figures.

The standout case revolves around the Adirondack, Empire Service, Ethan Allen Express, Maple Leaf, Vermonter routes, which all appear to operate with exactly the same passenger car consist of 5 Amfleets: 4 coaches and 1 cafe car. There does seem to be some non-obvious mixing of the individual carriages: some are high-capacity Amfleet I cars, while others are lower-capacity Amfleet IIs (the second-generation Amfleets were generally intended for long-distance service). But even accounting for that variation, some numbers are really outlandish.

My calculations based off of Amtrak's figures come up with an average of 554 seats per Maple Leaf, for instance, an impossible number for a 5-car train. On the other end of the scale, my math produces an average of 87 seats per Vermonter. It would seem that each of those trains has a typical capacity around 340, so something isn't right.

I'm not sure what's happening here, but it seems that Amtrak—typically operating fixed-size consists—is charging routes based on the number of seats the trains would have had if they were actually shrinking or growing trains in accordance with demand.

In some ways, I can see how that's not really a problem—4 out of 5 of those trains terminate in New York City's Penn Station (the last, the Vermonter in Washington, D.C.), where the added hassle of switching out train cars might gum up the works so much that it's better to just run with interchangeable fixed-sized trains. In the name of operating efficiency, some accounting tricks might not be a bad idea in order to keep the charge per route equitable. But it still seems to be covering up the true costs of running these trains while also inflating the overall cost of operation. Trains would need to be sized for the maximum load on any of the interchangeable routes (which in this case appears to be on the Maple Leaf).

Nationally, there seems to be a pretty consistent practice of under-counting seats by about 15%, though some routes work out to be very precise: Some state-run and mid-continent corridor trains seem to have pretty close numbers, and the Acela Express, with a fixed consist of 304 seats, is just about bang-on as well. My local long-distance route, the Empire Builder, is also low by a considerable margin for FY2011, though I believe that was due to a long string of annulments west of Saint Paul due to flooding which dropped ridership by about 12%. FY2010 numbers were almost a perfect match in comparison.

Anyway, here are the basic results of all this work (with further details in my spreadsheet):

Saturday, August 18, 2012

"Amtrak 90" plan

I just came across an interesting report called "Amtrak 90" put together back in 1982 which proposed a new route structure.   It is a much better mesh than what we have today, even featuring cross-border links to Winnipeg, Manitoba and Calgary, Alberta, though I consider a few links to be missing (Rochester, MN and Green Bay, WI, for example).  But then again, the plan only looked at an 8-year time horizon.

Anyway, some weekend reading.

Friday, August 17, 2012

Finding solutions for the Empire Builder's capacity problems

Amtrak passengers disembarking in Saint Paul.

Despite its unreliability and low frequency of service, Amtrak's Empire Builder often faces capacity issues. The train serves 45 stations along the way from Chicago to the Pacific Northwest, resulting in an immense possible number of trip combinations.  Some trips become impossible to book because of other riders who overlap on part of the route.  Despite the common wisdom, not everyone is traveling the entire 2,200-mile distance: The average trip is around 720 miles, taking about 15 hours.

In recent years, stations in western North Dakota have seen growing ridership because of the oil boom in the Bakken formation.  This past year, Williston—smack in the middle of the train's route—surpassed Minot to become the busiest station in the state of North Dakota.  While the numbers aren't huge (Williston sees about 82 boardings and alightings each day), it does make a significant impact on a train that's only running once daily in each direction and is making 44 other stops along the way.  Calls for extra capacity on the line have become common in recent years, and two weeks ago, Montana's two senators reiterated that request.

The fleet of locomotives and rail cars used by the Empire Builder is actually pretty large.  Because the journey between Chicago and the coast takes about 2 days, there are 4 or 5 sets of equipment in use at any given time.  The typical train is 2 locomotives plus 11 cars, and that grows to 12 cars for the run between Chicago and Saint Paul—an extra car gets added to the train for that segment because of typically higher demand between those large metropolitan areas.

Going by the consist listed on Wikipedia, the minimum fleet size is 67 vehicles, broken down like this (not counting spares for use during maintenance periods):
  • 10 GE Genesis locomotives
  • 5 baggage cars (Seattle)
  • 5 transitional crew/sleepers
  • 5 coach/baggage cars (Portland)
  • 15 sleepers
  • 17 coaches (3x per train plus 2x CHI-MSP cars)
  • 5 diners (Seattle)
  • 5 sightseer lounge/cafes (Portland)
Personally, I'd love to see the Empire Builder running two or three times daily, but considering that the going rate for each new passenger rail car is somewhere north of $4 million these days, plus the political environment of the moment, it's hard to see anyone plunking down the hundreds of millions of dollars required to double or triple the fleet size of one route, plus what may be hundreds of millions more for track upgrades along the 2,200-mile distance.

(It is a lot of money, but the train fleet listed above is probably cheaper than a single Boeing 747.  A new individual set of equipment probably has a value around $55 million.)

Other options involve adding extra cars to each existing train.  These wouldn't help solve the problems with on-time reliability or frequency of service, but they would help alleviate capacity problems.  Amtrak always claims that they don't have much equipment to spare, so it might be an issue to lengthen the Builder.  Since there are 5 sets of equipment needed for regular operation, adding one car per train means you must find 5 rail cars.  This really shouldn't be a problem, but Amtrak's acquisition of new equipment has been very bursty over time, with new Superliner cars only being built about once a decade.  There are only 430 active Superliners across the country according to this roster, and they're divided among about a dozen different routes.

Rather than finding 5 cars to lengthen every train, it might be possible to find 1 or 2 cars to run on just part of the route.  The extra car running from Chicago to Saint Paul could be extended to run to service stops in Minot, North Dakota (with one additional car acquisition) or Havre, Montana (with two).  This seems like the cheapest and simplest option, and could be scaled up to become an extra car over the entire route over time, possibly by restoring some rail cars previously damaged in crashes (costing $1 to $2 million per car rather than the $4 million for new equipment).  Hopefully that could be implemented in the short term, but it might take time to acquire equipment and negotiate the switching operations with BNSF (looking at track layouts, it appears that adding/removing a car would be easier in Havre than Minot, but either location might require track work too). 

But another option which would free up capacity on the Empire Builder while also improving reliability for many users would be to add more trains in shorter corridors along the route.  Amtrak is currently studying a second train between the Twin Cities and Chicago, with a report due around February 2013.  We can estimate what the report might conclude because a second daily train was previously studied in 2008–2009, though that was for the restoration of the North Coast Hiawatha all the way to Seattle.  That report suggested it would cost around $47 million to improve the line between Chicago and Saint Paul, and there would be other costs for purchasing equipment and training crews.

The first things that leap to my mind when I think about new train cars for this corridor are the two Talgo trains that were being built for Wisconsin's Hiawatha train between Milwaukee and Chicago.  The contract for those was canceled in a fit of political shenanigans last year, but they still sit, waiting to be used.  The pair only cost $47.5 million, though there are additional costs: They need locomotives, a maintenance base, and a contract with Talgo to maintain the trains for the next 20 years.  In all, a second train to the Twin Cities will probably cost $150 to $200 million.

We'll see if Amtrak's estimates come out in that range or not—the nature of freight traffic has been changing along the route due to dropping coal traffic and increasing movements of oil and frac sand, so costs could go up or down.

Giving riders on the busy eastern segment of the Empire Builder route another option should free up seats for people who do want to travel west of the Twin Cities, which will hopefully help residents of western states who rely on the train to get around.

Friday, August 3, 2012

Untangling the busiest rail chokepoint in Minnesota

A Friends of the 261 excursion train crosses the lift bridge in Hastings. A second bridge is proposed here if passenger traffic increases.

We in the Twin Cities are underserved by intercity trains. In a region of 3.3 million people, we only have Amtrak's Empire Builder stopping once each direction each day as it plies its way between Chicago and Portland/Seattle. However, this is still a significant hub of rail activity. Transcontinental lines of two railroads connecting the Pacific Northwest to Chicago mingle here. The Empire Builder uses both—BNSF Railway's route to the west, and Canadian Pacific's line to the east.

One of the great mixing bowls for freight traffic in our region is located just east of the Saint Paul Union Depot. Up to 150 mostly long, slow trains run through or shuttle around the area on busy days, moving about 10,000 rail cars in the process. The twin lines of BNSF and CP actually share track for a few miles before criss-crossing each other in Newport and again just north of Hastings. There are three significant rail yards in the area, one each for BNSF and CP, and another from Union Pacific, just to keep things interesting. BNSF and CP also each have big offloading facilities for automobiles along the tracks—the former near Dayton's Bluff and the latter in Cottage Grove.

Major rail traffic flows in the Twin Cities.

It's a busy area, expected to get even busier in the coming years. The freight railroads are expecting to see traffic go up by 3 to 5 percent annually (though some of that traffic increase comes from lengthening rather than adding trains). Proposed passenger services (5x daily roundtrips for the Red Rock commuter service and 6x daily roundtrips for higher-speed train service to Chicago) would also have a significant impact. Some additional tracks will be needed in the area, plus possible commuter train platforms. This turns out to affect a lot of different things along the corridor, and so the Ramsey County Regional Rail Authority partnered with the railroads and several other organizations to plan things out with the East Metro Rail Capacity Study (currently available in a draft form).

The study has identified dozens of possible improvements in the corridor from downtown Saint Paul to Hastings on the east side of the river where passenger trains run, plus a few things on the east side also thrown in to help with overall traffic flow.

A big problem in the area is speed: Trains often slow down to a crawl of 10 miles per hour or less as they switch between tracks, cross bridges, and climb some significant grades. Tracks also get blocked for extended periods as trains move in and out of yards. One of the first improvements recommended in the study is the replacement of 50 to 60 existing low-speed switches with ones capable of moving freight at 40 mph, which alone should make for significant differences in capacity. Mile-long trains whit may have taken 6 minutes to move past a point should then be able to pass by in 90 seconds.

Speeds are also planned to be improved throughout the corridor by realigning existing rails to even out curves which currently require slowing.

Another issue has been trains blocking main tracks as they serve yards and other facilities adjacent to the main line. In particular, Canadian Pacific's automobile offloading site in Cottage Grove has a siding that's too short, often leaving the head end of the train sticking out on the main track. New or lengthened sidings are planned in that spot and in several others to help keep the through tracks clear.

Now, goods can tolerate a fairly significant amount of delay as they move by rail, but people are much more finicky. In order to provide the best service, a couple of flyover bridges for passenger trains are outlined in the plans. These sound exotic, but it's worth noting the presence of several major road bridges in the immediate vicinity. The first flyover is planned just east of the Saint Paul Union Depot and would skip over tracks at the Division Street wye and drop trains right into an area known as the Hoffman Interlocking. The bridge will be relatively steep at a 2.5% grade (2.76% accounting for curvature), so it could only be used by passenger trains. Heading southeast, the flyover would lead straight into a dedicated passenger track.

(It's worth noting that a tunnel had also been proposed here, but that idea was discarded because it would have put the tracks below the water table and well within the 100-year floodplain of the Mississippi).

Proposed Saint Paul Union Depot flyover. View East Metro Rail Capacity Study in a larger map

Another flyover is planned down near Hastings where the BNSF and Canadian Pacific lines split in order to traverse opposite sides of the river. This would lead to a new lift bridge next to the existing one, which would be aligned to direct trains straight to the west side of the existing Hastings depot.

The last big operational issue I'll mention is the crossing over of traffic between BNSF and CP at both Newport and again just north of Hastings. There's little need for this traffic to cross over twice, and one major change put forth in the plan is a relocation of BNSF's existing tracks down along the river to a new alignment next to the CP tracks which closely parallel U.S. Highway 61. This would move the tracks out of the Mississippi River's 100-year floodplain and open up the opportunity to use the old rail alignment as a recreational bike path.

It's not clear whether all of this will be needed, so it's important to note that many improvements will either move forward or be put on the back burner on a case-by-case basis. However, it was an important exercise to go through, since the changing track layouts affect other projects in the area. The Union Depot flyover has impacts on where the proposed Bruce Vento Trail bridge can be situated, and other changes farther down the line affect things like the placement and type of support columns for the replacement Warner Road bridge over the tracks (now planned to be built of steel rather than concrete), and siting of potential Red Rock Corridor stations (the study recommends changing the proposed platform locations for both the Lower Afton Road and Cottage Grove/Langdon Village stations).

How much traffic growth will actually occur remains to be seen. The study planned for an overall increase of 36%, but that value isn't tied to any particular year in the future. But certainly some growth will happen, and it's important to be prepared.

Area transportation planners learned from their experiences with the Northstar commuter line that it was important to get the host railroads involved in discussions early and keep them involved. Past efforts at computer modeling had not gone over well since the railroads didn't know what assumptions had gone into them. This time around, the study group interacted with the railroads frequently and tried to get as much detail about current operations that they could.

The overall price tag for the proposed improvements is projected at $827 million, though transportation officials estimate that about 2/3 to 3/4 of that will be paid for by the railroads themselves, since many of the changes are needed by them regardless of whether additional passenger service happens. It's likely that this will be divided up into a number of smaller projects as need arises and funding becomes available.

If and when—hopfeully when—passenger traffic increases on the line, riders will be assured a much less delay-prone ride.