Mathematically, the London Underground is too fast. No, really.
To get to my desk this morning, I had to travel through 14 London Underground stations. If I had to pick one phrase to describe this journey, “lightning fast” wouldn’t even make my shortlist.
However, anew study from Dr Marc Barthelemy, a statistical physicist at the CEA research centre in Saclay, France, says that the London Underground is actually too fast for the capital. In fact, the speed at which Tube trains travel (around 21mph on average) is causing bottlenecks, and should theoretically be slowed to make them marginally faster than cars in the city.
“The optimal speed to keep things running smoothly would be around 13mph – 1.2 times the average road speed.”
The thrust of the argument – illustrated using network models – is that bottlenecks outside the city centre slow everything down, because that’s where everyone changes modes of transportation. The optimal speed to keep things running smoothly would be around 13mph – 1.2 times the average road speed.
The computer model they built worked purely for Underground stations and roads. “We create these connections, and then we make an assumption, which is: when someone wants to go from A to B, they look for the quickest path – whatever the mode,” Barthelemy told the BBC. The Underground, it turns out, decreases congestion in central London, but increases it around the edges.
“The fact is that these networks are coupled with each other. Optimising something on one network can bring bad things on another network,” he explained.
“Downtown Manhattan is so full of grumpy commuters that it needs the fastest trains it can get.”
So do all of the subway systems around the world need to chill out a bit? Not so fast (pun intended), says Dr Barthelemy, who also created an identical model for the New York Subway – with completely the opposite results. “Surprisingly enough, the network in New York is much more centralised than the one in London.”
This means, in practice, that Downtown Manhattan is so full of grumpy commuters that it needs the fastest trains it can get “even if that increases the congestion at some peripheral points – the entry points to the subway.”
So how fast is the London Underground compared to other underground train networks? Pretty average really:
|City||Average speed||Top speed|
What quickly becomes apparent when looking at the data is that top speeds are all quite similar the world over. The average speed is more significant, and that comes down partly to the age of the system, but also to city planning specifics: how straight are the tracks? How many stops are there? How great a distance is there between the stations?
In London, for example, the fastest line is the Central line, with the Victoria, Jubilee and Metropolitan lines close behind.
“Is a faster train network any use when the rest of the country is one giant bottleneck?”
I wouldn’t expect slower trains to be introduced by TFL anytime soon – not least because, as an announcement goes, it’s hard to spin, even if you have mathematical models in your favour. But even if it is 100% correct, it certainly needs more assessment, as Professor Michael Batty from University College London told the BBC, “It really is just a network model. There are no capacities on the network – it’s not really a flow model, like the ones that Transport for London actually use.”
Still, food for thought as HS2 is still being debated: is a faster train network any use when the rest of the country is one giant bottleneck?