Believe it or not, there exists a Nash equilibrium for this. There are several papers discussing various approaches to optimally distributing traffic so everyo

Believe it or not, there exists a Nash equilibrium for this. There are several papers discussing various approaches to optimally distributing traffic so everyone gets the best possible route given all the cars on the network: http://dl.acm.org/citation.cfm?id=2008646 http://dl.acm.org/citation.cfm?id=2008646 http://www.sciencedirect.com/science/article/pii/S0191261504000517 http://www.sciencedirect.com/science/article/pii/S0191261504... http://ieeexplore.ieee.org/document/1181966/ http://ieeexplore.ieee.org/document/1181966/ I have read a paper (which I can't find now) that claimed that cooperative optimization had visible effects on overall routing with as little as 7% of the traffic population participating in the cooperative routing scheme. So it doesn't even require that everyone use the same scheme to get some benefits, which is nice.