Thursday, October 14, 2010

Understanding Braess

And calming traffic.

Simply put - adding capacity to a network decreases overall efficiency.

See Scientific American: Removing Roads and Traffic Lights Speeds Urban Travel

Why? There's math for that (see Wikipedia). How I see it is based on the psychological experiment that asked people if they'd rather live in a neighborhood where they made $100,000 and their neighbors $50,000; or a place where they made $200,000 but surrounded by people making $400,000. People overwhelming wanted to be richer than their neighbors even when that meant passing up a doubling of their income. A study from the U.K. confirmed this experiment: Happiness Is ... Making More Money Than the Next Guy. Or as They Might Be Giants sang, "I don't want the world, I just want your half".
This is how drivers behave - why they do not make the most efficient choice. Some drivers are simply looking for a way to get to the next spot before the other guy.

You see this as drivers speed down Lake Ave past Oryana and McGoughs. They aren't going to save a lot of time by doing this but if they can get ahead of the car that they were behind a few minutes ago on Fourteenth then they are satisfied.

And as an experimental guy found out in Seattle, one bad or good driver can impact everyone else for the worse or better.

I believe this is the essence of the Braess paradox, additional network choices lead to inefficient decision because the agents look at only getting to the next hop in the network. By eliminating choice you increase efficiency since there is no competition for an alternative route.

This has been proven New York, Stuttgart, Seoul, Los Angeles. It can work in Traverse City too. Plus, it is much cheaper to close a road than to build a new one.

This is something I want Traverse City to consider as the City plans to introduce traffic calming ideas. I'm going to listen to them then will have more to say about it and about the proposed Boardman Lake Ave.