So if unit trains are more efficient then loose carload shipments, why not run short unit trains? Would not a 10 car unit train from point a to b and back be efficient?
Two reasons: crew wages and locomotive efficiency.
Now if you talk about a 50-car train with only one crew member, then maybe.
Ten cars is small enough to be handled as a block on a larger train. Something that small is not a major improvement over single-car shipments. 50 cars, as suggested above, even with a 2-man crew, seems to be about the minimum for a unit train.
Basically, unit trains save money by improving equipment utilization. They also don’t need space in yards, switch engine time, etc. But they also add costs when they require extra road crews, cause MofW forces to shut down and let them pass, take up dispatcher resources, etc.
Their economics are a trade off between the costing elements. I’ve got an AAR study on short haul intermodal that uses:
- Train Mile costs - $3.53/mile
(dispatching, interferance with MofW and other trains,etc) - Crew Costs - $8.30/mile
That’s $11.83/mile for those two elements alone. For a 10 car train that would be $1.18/car mile. On a 50 car train, it would only be $0.24/car mile.
A 10 car unit train can make econonomic sense, but it would have to produce enough in savings in some areas to offset its added costs in other areas. A 10 car unit train operated with a one person crew from an ethanol plant on a light density branch would have almost no “train mile” cost and reduced crew costs.
Will it happen? It depends on the union negotiations. If the crew costs stays up the loads will probably move in regular locals.
I have brought this up before,. NS currently runs unit “bottle trains” of molten steel from Northwest Indiana to South Chicago. So, it can be done.
ed
Wow, every now and then I really learn something on this forum (not that I know that much to start with). Really excellent analysis by greyhounds.
I was kind of thinking along the lines of MP173.
If NS can run 8 car mixed train and make money on its busy Decatur IL to St. Louis Missouri line, why can’t a 10 car unit train make money?
Gabe
The “bottle train”, which was originally an Interlake Steel intraplant operation between its blast furnace and its finishing mill, is only slightly less specialized than the movement of a loaded Schnabel car. The shipper supplies everything but the motive power and crew, the cars are very heavy and are probably covered in the Special Instructions, and the haul is within the terminal area boundaries. I would also assume that ISG or Mittal pays a premium rate for this service. It may be a unit train operation of sorts, but it is too specific to a given area to be held up as an example for elsewhere.
Actually, the small unit train concept is one that has great profit potential for the railroads, yet has hardly been exploited due to hard wired misconceptions regarding the “efficiencies = profits” mindset. The small unit train concept is a perfect fit for short haul markets.
Assuming Ken is probably correct in his assessment, the per car cost for a 10 car unit train is about $1.18/car mile, which is a much higher cost than the longer trains. However, since the shorthaul market is one which the railroads have all but ceded to truckers, it is the truck cost structure that dominates rates in such corridors. Trucks probably can’t handle a cost per mile lower than $1.50, and are probably more in the $2.00 - $2.50 range. On a ton mile basis, the 10 car short haul rail shuttle cost comes out to about $0.01/ton mile, while the trucker’s ton mile cost is about $0.04/ton mile. Thus, there is a pricing opportunities for railroads in which the railroad can charge a higher rate than the usual long haul rates, yet still get the business because the higher rail rate still beats the trucker’s break even rates.
The key of course is to haul a commodity that allows the train consist to remain intact, and to provide the service at a consistent availability interval so that timing constraints of the shipper don’t force the traffic onto trucks as a last resort.
As for long haul, railroads do occasionally add 10 or so car sets to regular manifests and drags if the space is available.
Actually they come by my house but in the form of 100 car manifest with blocks of 10 cars or more included. When they get to where they are going they become blocks or singles for locals.
Wisconsin Central (and now CN) have been doing this for at least eight years, running 25 car unit trains of sand and gravel from quarries in Sussex and near Slinger, Wisconsin to dealers in Illinois, and they seem to be making money off of it.
In the early 1970s , RDG ran some short distance unit trains with reduced crew size, etc. They called it the BEE Line. Some of their freight units had the Bee Line logo on them. That all stopped before Conrail.
Randy,
Do the cars belong toCN, or is it a leased package.
Georgetown railroad down here leases the entire train, sans crew and power, to UP.
Lets them run pretty much whatever size train the aggerate buyer wants.
This may or may not apply, but the Wisconsin Central kind of used that philosophy when they started in the 80’s. Short, fast trains. It worked very well for them. I don’t know that they were unit trains, however.
I remember reading in Pacific Rail News in the early 1990s about SP experimenting with a sprint train (I think defined as about 12 cars) to run perishables back east. I have never heard anything about it again. (Sarcasm alert) I wonder why.
When the Coast Line was closed due to slides in early 2005 the Oil Cans detoured through the valley. Initially they had all 78 or 91 cars. However, after the first few times they cut it down to 26 cars. I have never been able to figure that out.
I think your off the mark on the 10 cars but the FEC was running 3 trains a day north and south. “We switched to eight trains a day (N&S) or more and were able to double our revenue ton-miles.” quoted from W. L. Thornton FEC Chairman.
Actually, I think Dave is misinterpreting the numbers I provided. I gave only two cost elements, “train mile” and “crew”. These were examples of “added” costs incured by the operation of an extra unit train as opposed to adding the business to existing freight trains. My point was that these added costs have to be offset by the efficiencies of unit train operation. (ie. improved equipment utilization, no switching
Another aspect of this is congestion. In general the short train takes as much track occupancy time and dispatcher effort as a long one. If the route is congested adding a train will impose delay costs on that train and probably every other train it encounters. In practice this means that short unit trains will usually be associated with low density lines where congestion costs are nil.
The steel bottle moves are doubtless made at high enough rates to cover all costs of the move. On a per car basis the rate is probably outrageous.
Mac
Way back in the late 1960’s, TRAINS had a lead-off article in “News & Editorial Comment” about Reading’s Bee Line Service. It was not unit trains but a short to very short haul truck-competitive service with negotiated variations in existing labor agreements regarding crew sizes, operating boundaries, etc. RDG determined that they could make money on this arrangement on hauls as short as 2 miles.
I just remembered that ADM is running a service at Decatur in conjunction with NS and perhaps CN of short haul grain trains from the fertile fields of Central Illinois to the “Supermarket of the World”
So, it can be done. I would think the key here is ownership of the cars. The asset costs per day would really chew up the railroads, if they were not careful.
ed
Yes, CN does own the cars, they are taconite cars released from ore service in Canada. Vulcan Material likes them for their fast dumping ability in freezing weather. (They have been known to run trains all through our Wisconsin winters.)