"(PhysOrg.com) – If you’ve noticed that Dutch trains experience less delays or that waiting times are shorter on the Berlin underground you can thank a team of European researchers whose advanced algorithms are optimising rail services.
In what has been described as a breakthrough in the field of operational research (OR) and has garnered a prestigious award, the researchers succeeded in developing advanced algorithms able to optimise planning and scheduling in vast, complex rail networks.
In practice, their work means that railways can be more confident that their routing choices will allow more trains, more passengers and more goods to safely traverse the same infrastructure while increasing punctuality, passenger satisfaction and operator profit. It’s a win-win for everyone."
After a quick perusal of the article, the use of these algorithms seems to have more merit and substance to it than my skeptical mind first anticpated. It will be interesting to see a more detailed explanantion of how it works, and what any informed critics have to say about it, either pro or con, strengths and weaknesses or limitations, etc.
Algorithms rarely allow for broken knuckles, drawheads, burst air hoses, crossing accidents and trespasser fatalities. All of which are too common on today’s US railroads.
[955 AM Weds. 05 May 2010 - Fixed the formatting- PDN.]
That’s exactly what I thought. But they at least “talk the talk”. From the linked article:
““It’s a proactive approach. You don’t have an optimal offline plan that can’t handle disruptions. You have a near-optimal plan that can handle as much disruption as possible,” Zaroliagis explains. Online planning, on the other hand, takes a reactive approach, dealing with disruptions as they happen in real time.”
“Although robust and online optimisation methods have previously been developed for other sectors, such as the aviation industry, they turned out to be inadequate when applied to railways due to the complexity and sheer size of rail networks. Whereas changing the take-off order of planes at an airport may cause notable disruption elsewhere, in a railway system, because trains have to use the same tracks, delaying the departure of one train can cause chaos across the network.”
I wonder is this was discussed at last week’s ASME/IEEE Joint Rail Conference at the Univ. of Illinois - Champaign-Urbana ?
low-profit/high-prestige/high-horsepower/fuel-guzzling container-trailer trains with
a number of higher-profit/under-powered/fuel-efficient manifest and grain trains,
the latter always impacts the performance of the former in unpredictable ways.
And if one of those minimal horsepower trains looses the pulling power of a single unit because of radiator water leaks, low oil pressure, clogged fuel filters, an electrical component failure, ground relay problems, running-out-of-fuel, or a power assembly goes “piston-in-prarie,” the next thing you know that crew dies on the law and it takes another five following crews down with it.
The real world of North American freight railroading ain’t no Lionel train set, that’s for sure.
And these trains cover short distances (10’s of miles) and heavily subsidized systems. The U.S. Railroads use flow forecasting models to schedule trains and crews. Savings from these models are significant. Even though railroads have problems, the models can accomodate a slow or failed train and prevent other trains from leaving terminals until there is a time slot for the new train.
That’s true in a lot of Europe, but in prime freight areas the opposite can be true. If you draw lines from the ports of Antwerp, Rotterdam, Bremen/Bremerhaven, and Hamburg, to the Ruhr, and then down to Milan in Italy you will find that freight outnumbers passenger. And on the old mainline down from Hamburg through München to Verona, except in the immediate München commuter area. In many German cities there are freight bypass lines that keep freight out of the downtown areas. In the Ruhr for example there are dedicated double-track freight only lines to keep freight and passenger movements from interfering with each other. One thing Europe really has is in a lot of cases there are 3 or more routes between any two cities, the only real exceptions are where you need to cross the Alps.
My favorite example of European infrastructure is the Dutch mainline through Barendrecht. The lines here consist of the double-track Highspeed line from Antwerp towards Amsterdam, the four track classic mainline used by lesser passenger trains and those freights bound for Amsterdam, and then a triple-track line solely for freight traffic bound for the Port of Rotterdam, giving you a nine-track wide mainline. This nine track section runs from a massive flying junction at the north end of Kifhoek humpyard (which sorts all the carload traffic for the Netherlands), to another major flying junction where the Port freight line swings west into the massive port of Rotterdam.
Bob’s last line sould have a place in the ‘Annals of Understatement!’
And BaltACD’s right on target as usual, as well.
Wonder where Forum member, ‘Marcimmeker’ is on this? If i’m not mistaken he lives just off that Dutch Rail Corridor and should have someing to say, he had posted a
Sounds like a good thing to me. It all depends on what you want the algorithm to do. NS’s blocking system is algorithmic and the soon-to-be train dispatching movement planner is algorithmic, as well. They can’t predict behavior, but they can react to it, so when you get a broken knuckle, your operation adjusts to minimize the overall impact on the system.
Shorter and by necessity more frequent. 1800 - 2400 ft. is the common length, although France and Italy have some areas with shorter limits. There is some movement in Germany towards trains of up to 1.5 kilometers on certain primary corridors. The European freight operators have seen fit that if they have lemons they might as well make lemonade. The shorter train lengths (and hence fewer cars per train) means that is much easier to make unit trains for commodities that you would never find in the US. Solid trainloads of Newsprint from Scandinavia to a distribution center in the Ruhr, or near Milan. Trainloads of bottled water for Evian from Eastern France to a distribution center north of London. Trainloads of structural steel from the Ruhr to a distribution center near Milan. And the biggest of all Refined Petroleum products to many, many distribution centers.
Humpyards absolutely, there are ten in Germany alone, and they have all the bells and whistles, plus there are lots of mini-humpyards so that switch crews don’t have to kick cars. Because of shorter train length most trains are only one block. So Maschen Yd. near Hamburg will send out one to three carload trains to all nine other humpyards, various portions of the port, and the Netherlands and Scandinavia six days per week (Europe takes Sunday off).
There is less Intermodal in Europe, but it is growing at similar rates to that in the US. Way more chemicals move by Intermodal in Europe than in the US, so 20ft. Tanktainers make up a far higher proportion of traffic in Europe than here. Also they have many more types of Intermodal Equipment, plus of course Rolling Highway (which we have no equivalent).
While algorithms may be designed for a perfect world that doesn’t really exist, they do serve as an ideal which can be used as a reasonable goal in the real world and can provide some direction for decisions by management. They should be considered as a useful management tool.
![http://www.swissworld.org/index.php?eID=tx_cms_showpic&file=fileadmin/user_upload/images/sisimg20060430_6671795_1.jpg&width=1024&height=1024&bodyTag=<body%20style%3D"margin:0;%20background:%23fff;“>&wrap=<a%20href%3D"javascript:close();”%20title%3D"Click%20to%20close%20the%20window">&md5=2fb23099aaf53bedfa7a53dc3f3adf38[/link]](http://www.swissworld.org/index.php?eID=tx_cms_showpic&file=fileadmin/user_upload/images/sisimg20060430_6671795_1.jpg&width=1024&height=1024&bodyTag=%3Cbody%20style%3D%22margin:0;%20background:%23fff;%22%3E&wrap=%3Ca%20href%3D%22javascript:close();%22%20title%3D%22Click%20to%20close%20the%20window%22%3E&md5=2fb23099aaf53bedfa7a53dc3f3adf38%5B/link%5D){kind=link}