Skip to main content



The relationship between traffic and roads

It is quite intuitive that building more roads is a way to solve traffic jam problems. As traffic flow can be separated into many different roads, there will be fewer cars in each of them. However, during the studying of modeling network traffic using game theory, we encountered a case where building an extra road might lead to more travel time.

Screen Shot 2015-10-16 at 10.16.15 PM

Screen Shot 2015-10-16 at 10.16.26 PM

In this example from the book, if there are total 4000 cars that want to travel from A to B, the travel time will be 65 minutes total at equilibrium.

However, if a more efficient (only cost 0 min to travel) road is built between C and D, the travel time at equilibrium becomes 80 minutes for everyone. As the textbook mentioned that this phenomenon where “adding resources to a transportation network can sometimes hurt performance at equilibrium” was known as Braess’s Paradox. This is somewhat caused by the self-interested behavior by all the drivers: if all of them wish to travel through the “fastest” road, then the “fastest” road becomes slow eventually.

The article (link 1) I found from WEIRD talks about a similar issue of building more roads might actually make traffic worse. Both the textbook and this article mention the story of Seoul, Korea, where the destruction of a six-lane high way actually improved the overall traffic through the city. It introduces a concept called induced demand: increasing the supply of something might make people want that thing even more, which sounds similar to the self-interested behavior mentioned in the textbook. This concept also reminds me of “the Tragedy of the Commons”: because people acting according to their self-interests independently, some common resource (the road in our case) becomes depleted and therefore, the best interests of the entire group cannot be satisfied.

However, I found this article a little bit biased. The author states that recently social scientist have collected enough data to show this phenomenon (of more roads leading to worse traffic) happens “pretty much every time” a new road is built and the evidences seem not sufficient. He only related to the study from two economists, Matthew Turner of the University of Toronto and Gilles Duranton of the University of Pennsylvania.

Searching about this topic even more, I came across another interesting article (link 2) on the CATO Institute that specifically goes against the WEIRD article. It opposes the “roads-induce-demand” but agree with the congestion pricing advocated by Turner and Duranton. Reading both of them gives me a deeper understanding of the complexity of traffic system. In our real life, under some correct combination of conditions, more roads can lead to more problems. However, designing and optimizing a transportation system is really complicated problem, and modeling of traffic equilibrium is only on small part of solving it.

 

http://www.wired.com/2014/06/wuwt-traffic-induced-demand/ (link 1)

http://www.cato.org/blog/debunking-induced-demand-myth (link 2)

Comments

Leave a Reply

Blogging Calendar

October 2015
M T W T F S S
 1234
567891011
12131415161718
19202122232425
262728293031  

Archives