Monday, September 22, 2014

Transit and congestion, part 1: is transit a solution to congestion?

The issue of transit and congestion comes up quite often. Oftentimes, transit proponents will sell transit to car drivers as a solution to congestion in order to get them to support spending money on transit solutions.

In reality, transit rarely allows for congestion to be reduced. The simple reason is that if transit does manage to get traffic down, the reduced congestion is likely to lead to induced demand as more people will make choices that result in them driving across the congested chokepoints of the road network... or maybe even transit users who have cars will see that congestion is light and so will switch back to cars!

Indeed, transit in most cases is actually much slower than cars, in North American cities, it isn't rare for transit to take 2 to 3 times as much time as cars to get anywhere, at least without congestion. Subways in dense cities where cars are forced to drive slowly are one rare exception. The only way for transit to be really attractive is thus for cars to see their average speed fall because of congestion. The moment when car drivers switch for transit is when they see they have an advantage to do so, and the most obvious advantage is time.

Average speed of cars and transit (with or without right-of-way) on an highway as it congests
What the previous graph shows is that the moment when people are going to flock to transit is when the speed curve of cars go down to the level of the speed curve of transit, which only happens in the case of congestion. Transit with a right-of-way, which may be simply buses with a bus lane allowing them to bypass lines of stop-and-go traffic until they reach the chokepoint of the network, will also be much better at it. Transit that go in mixed traffic will also see speed fall because of congestion, and people will go to them much less readily.

So, is transit useless then? No, it is very useful to avoid having central urban areas be suffocated by traffic. 

Let's take a simple representation of a CBD and a suburb, connected with a single highway with a capacity of 20 000 people in cars over an AM peak period (6 to 9 AM). Currently, the road is at capacity in the morning peak, when most people commute to work.

The suburb is predicted to grow 10% over the next 10 years. This growth in population is likely to result in more jobs too, so normally you might see the CBD grow likewise to provide jobs to these people, because there is an advantage in industries to stay close to similar companies.
However, this would result in more people wanting to go through the road than it can bear. Congestion will rear its ugly head, then people will try to find ways around it. One of the ways would be to live in the CBD, but for many reasons, it tends not to be easy to do. Or roads can be widened, but that is often supremely expensive in urban areas. One of the most likely alternatives that may occur is that instead of being in the CBD, the job growth will occur elsewhere, in an industrial park near the suburb.
So, in this case, the CBD stagnates while suburbs and job centers in suburbs grow. This may not seem so bad, but this tends to lead to downtown areas losing vitality and importance, in the end, the CBD may even start losing jobs to industrial parks in suburbs. The new developments also tend to be car-dependent, forcing people to own cars to get anywhere, which increases costs for everyone.

Transit can offer an alternative. Indeed, transit can increase the capacity of roads, a single bus can carry 50 or 60 people and take the place of just 2 cars on the road. Betting on transit is a cheap way to increase road capacity and so to keep the CBD growing and staying a vital part of the region rather than becoming just another industrial park amongst others.

This kind of transit is a mixed blessing... yes, it allows the central city to keep growing and keep its relevance, but it also enables residential-only suburbs to keep growing and sprawling. So it prevents job sprawl, but not residential sprawl. It can even be seen as an enabler of sprawl, as a mere aspect of a road-only policy.

So transit doesn't reduce congestion... but allows more people through nonetheless because of its greater spatial efficiency. But there is another aspect of the relation between transit and congestion that needs to be talked about, and that is the disastrous effect congestion has on transit. For transit has actually historically been a victim of congestion.

It's late and it's a distinct enough subject to warrant another post, so I'll stop here tonight.

1 comment:

  1. The speed of cars also depends on whether they have a grade separated or at least mostly-unimpeded ROW too. In general, transit will be slower than cars on the same class of ROW, especially if transit has to run in mixed traffic. But in a city that has subways and not freeways, subways might actually be faster. Comparing travel times on the LA Red Line to Google's estimates of driving time (without traffic) shows that the subway generally wins, and not by a small amount either. It's only the presence of freeways that allows cars to be faster than rapid transit.

    ReplyDelete