Passenger Trains > Amtrak's new Chief Engineer-Change or Business as Usual

The news about Mr. Pohlot's appointment raised some questions I've tried to articulate following. I don't
profess to have knowledge of Amtrak's inner workings; this is just the way I see it.

Amtrak's engineering dept. includes 3,500 of 19,000 total employees and accounts for about $1B of expenses.
It's primary mission is to maintain the NEC infrastructure.

Next is the mechanical dept. with 4,500 employees at sites around the country. It maintains and
repairs locomotives and rolling stock.

Largest is the transportation dept., with about 8,400 people. This includes T and E, OBSC and support
personnel needed to operate trains.

Of these three, the last should be most controlled by work rules, number of trains operated,
similar to any railroad. These are nearly fixed costs, and present the least opportunity to
cut costs by improving efficiency.

On the other hand, the first two are the best places to cut waste, fraud (overtime), corruption, etc.,
and are likely where most of these are hiding. It was almost comical watching Mr. Boardman explain
Amtrak's extraordinary overtime costs to a clueless congressional committee. Definitely not from
an MBA textbook. The IG has uncovered improper procurement practices, involving hundreds of
millions.

When forced to make cuts, Amtrak's response is to cut #3 by eliminating trains, waiters, etc.
as more "publicly visible" and often politically unpopular tactics. This has worked a number
of times, causing cost-cutting politicians to back off.

The engineering and mechanical departments never show up on the radar, yet they alone account for over
half the $4B expense budget and 42% of the work force. I find it difficult to believe it costs $1B/yr
to maintain a few hundred miles of RR, or the mechanical costs for the CZ are $45,000/day.

Sound like some excellent opportunities for reorganization or privatization.


I wish Mr. Pohlot the best, and hope he can make Amtrak a better company.

jbaker Wrote:
-------------------------------------------------------
I find it difficult to believe it costs
> $1B/yr to maintain a few hundred miles of RR, or the
> mechanical costs for the CZ are $45,000/day.
>
> Sound like some excellent opportunities for
> reorganization or privatization.

All of that few hundred miles is electrified (some on the south end with caternary dating back to the 50s), all of it is Class VI (125 MPH or better), much of it 3-4 tracked, and much of the trackwork done in darkness when the NEC is quieter -- with associated inherent inefficiencies. The longest bridge (over the Susquehanna) dates to around 1914, I believe, and many others have similar birthdates.

I'm not saying there aren't opportunities, but there are also challenges....this is NOT the BNSF or UP operation in the Northeast.

As a consultant to railroads for nearly 20 years, I did a lot of comparisons of the costs of maintaining permanent way. I found it most convenient to boil everything down to a cost per 1,000 gross ton miles (GTMs), since much of the cost of maintaining track (but by no means all) is driven by traffic density, as measured in gross tons per track mile.

US and Canadian railroads (Class Is) all cluster around $1 per thousand GTMs. Commuter railroads (Tri-Rail, Metrolink, Metra, MNCR, etc.) are in the range of $3 to $8. Amtrak comes in at $14.

Now maybe it really does cost that much to maintain high speed track, but if you look at the Europeans, with a few exceptions they're all below $10 per thousand GTMs. Sweden is at $2.50.

Comments are welcome. Oh, and I made these comparisons several times, carefully selecting comparable MOW accounts. The Amtrak number does not include ET or signals.

Shoretower's expertise and input may shed some light on what has been a real mystery.
The IG could hire a business management consulting firm, with railroad expertise, to
evaluate the engineering and mechanical departments. This would include such comparisons
as mentioned, with systems in Europe and CR.

So when the next person asks why Amtrak needs a certain amount of subsidy, should the automatic
response be "It's the long distance trains"? It's politically convenient to say, and widely
believed, but is it the truth?

The engineering and mechanical departments account for considerably more expense than train
operations, yet they are quietly ignored; and then there's the union's part in all this.

shoretower Wrote:
-------------------------------------------------------
> As a consultant to railroads for nearly 20 years,
> I did a lot of comparisons of the costs of
> maintaining permanent way. I found it most
> convenient to boil everything down to a cost per
> 1,000 gross ton miles (GTMs), since much of the
> cost of maintaining track (but by no means all) is
> driven by traffic density, as measured in gross
> tons per track mile.
>
> US and Canadian railroads (Class Is) all cluster
> around $1 per thousand GTMs. Commuter railroads
> (Tri-Rail, Metrolink, Metra, MNCR, etc.) are in
> the range of $3 to $8. Amtrak comes in at $14.
>
> Now maybe it really does cost that much to
> maintain high speed track, but if you look at the
> Europeans, with a few exceptions they're all below
> $10 per thousand GTMs. Sweden is at $2.50.
>


Curious....what are the figures for Germany, France, and Switzerland?

Thanks!

In fairness, Amtrak MW etc crews operate around a fairly dense schedule of operations.

In a place like Jersey Avenue (milepost 34 on the NEC in NJ), you may get 15 minutes work in during an hour. Grab your time and track, lose it, and wait ten minutes for another 5 to 7 minutes. I'm not knocking the MW guys or DS or trains, but it's tough to keep costs in line when you have 15 minutes of work and 45 minutes of standing clear. Short of running trains against the direction of traffic for an hour or two at a time, and using the boards, etc this problem isn't going away as long as you're walking track in daylight during the week.

From time to time, I've had the opportunity to watch the reconstruction of the substation at COUNTY. I've never seen out and out goofing off, people sleeping in the trucks, etc. Never. But there is a lot of waiting around for somebody to complete a line or check something, so the next team can do their part of the project. Whether SKANSKA etc could manage the time and project more effectively isn't something I'm competent to consider.

(Although the continuing screwup at Lancaster Amtrak station does cast considerable doubt on Amtrak's project management, project supervision, cost control, etc disciplines.)



Edited 1 time(s). Last edit at 12/10/12 16:30 by Lackawanna484.

shoretower Wrote:
-------------------------------------------------------
> As a consultant to railroads for nearly 20 years,
> I did a lot of comparisons of the costs of
> maintaining permanent way. I found it most
> convenient to boil everything down to a cost per
> 1,000 gross ton miles (GTMs), since much of the
> cost of maintaining track (but by no means all) is
> driven by traffic density, as measured in gross
> tons per track mile.
>
> US and Canadian railroads (Class Is) all cluster
> around $1 per thousand GTMs. Commuter railroads
> (Tri-Rail, Metrolink, Metra, MNCR, etc.) are in
> the range of $3 to $8. Amtrak comes in at $14.
>
> Now maybe it really does cost that much to
> maintain high speed track, but if you look at the
> Europeans, with a few exceptions they're all below
> $10 per thousand GTMs. Sweden is at $2.50.
>
> Comments are welcome. Oh, and I made these
> comparisons several times, carefully selecting
> comparable MOW accounts. The Amtrak number does
> not include ET or signals.

Whoa Whoa Whoa this is absolutely the worst way to compare the spending.

How on earth can you draw a comparison between an freight railroads track maintained to FRA Class 4 and 5 to Amtrak's Class 6 track? And the commuter railroads cited aren't running Class 6 track either.

Second and more importantly dividing the cost per Gross Ton Mile is devious. The freight railroads run trains of 7 to 8,000 tons some as high as 13 to 14,000 tons. Amtrak runs trains of 700 to 2500 tons (Auto Train) the freight railroads rack up much higher Gross Ton Miles to spread the cost. Remember a GMT is one ton carried one mile, and includes lading tare weight of the cars and locomotives.

Also do your computations include the commuter railroads that are hosted on the NEC? How about the freight run on the NEC?

Where did you obtain your costs for the European countries? Was any adjustment made for axle loading differences? Due to freight use Amtrak has to maintain for 286,000 GRL, I wouldn't think that anybody else in the world is anywhere close

I'm no Amtrak apologist but this sounds like Representative Mica type of statistics.



Edited 1 time(s). Last edit at 12/11/12 14:17 by trainman630.

trainman360 has hit the nail on the head. Amtrak has a "perfect storm" situation that it needs to address in the maintenance: 1) the need to maintain at class 6 track, 2) the higher (much higher) axle loadings than Europe, 3) the short maintenance windows, and 4) the need to catch up on many years of deferred maintenance or obsolescence of bridges and other track components that was taken care of in Europe at a much earlier time (especially in the 50's and 60's). I'm sure that there is room for improvement (there always is on any railroad), but the numbers do not strike me as particularly out of line.

The cost per ton-mile as a wear index for rail and other track components is a fair way to compare rail replacement costs but not to compare the whole infrastructure operating cost.
The cost of getting labor and machines onto the track for an hour of productive work is the huge variable between dense passenger service lines and other lines which may have large tonnages but have time for work blocks between trains or fleets of trains.

And in some cases take tracks out of service for lengthy periods of major work while trains are rerouted. That's one way to work very efficiently in a way that cannot easily be duplicated on a line like the NEC.

I've always thought maintenance on high traffic, day use mainly, transit rail is performed
at night and on weekends. All ROW maintenance crews work a regular say, 9pm-5am shift, and weekends
not as overtime, but as normal work hours. I've never thought of railroading as an 8-5 job,
at least it never was in my family. Same goes for weekends and holidays, just another day.

Seems like multiple tracks would facilitate routing freight trains or other night traffic around
work areas.

Trying to do maintenance during the day in the NEC doesn't sound very efficient. What's wrong
with shutting down a track for a week or two, letting passengers endure some inconvenience and delays?
We put up with it all the time on highways.

Okay, the TO experts have weighed in. Let me make a few observations:

1) For five years I was responsible for planning and scheduling a $100 million annual track rehab program on the New York City Transit Authority, so I think I know a bit about the subject.

2) I introduced NYCTA to the idea of taking tracks out of service for trackwork. We usually could get five hours during the middays (outside lines) or five on the midnights (in the subway). For major stuff like switch and crossover renewals we had to do 55 hour weekend closures.

3) Production rates sucked. The gangs didn't get all that much done per day. But there really weren't other options. We carried too many people to use buses.

4) Amtrak differs from North American freight railroads in a host of ways, but it's often possible to take a track out of service for a long period (weeks or months) for major renewals. They just leave the P-811 sitting on the track at night, and work during the day. Most freight railroads don't have the luxury of multiple tracks and have to clear up every night.

5) I compared Amtrak with European railroads precisely because Amtrak is more like a European than a US railroad. The only segment of the NEC to carry significant freight is between Perryville and Baltimore (about 20 MGT per year). The rest is mostly local freight only.

6) But the fact remains that Amtrak spends $14, which is more than virtually all European countries, and much more than Sweden, which has high speed trains, and freight, and spends $2.50.

Some considerations for Shore Tower (I was in Shore Tower during the cutover to CTEC in 1992(February?) by the way):

Are you considering P-811 work maintenance? This is typically "capital" work for Amtrak.

A P-811 that is "cut-in" and it's large contingent of support machinery is probably the only thing that Amtrak can leave in place on the main line (except switch replacements which are typically weekends). And whenever they work, everyone and every machine has to stop work for passing trains. With the catenary overhead, it's not permissable to raise the track, which means you have to use more force to advance. Those two reasons are why freight RRs get more production out of their P-811 style equipment than Amtrak.



Edited 1 time(s). Last edit at 12/13/12 13:41 by RRTom.