A multiagent solution to overcome selfish routing in transportation networks

It is well-known that selfish routing, where individual agents make uncoordinated greedy routing decisions, does not produce a socially desirable outcome in transport and communication networks. In this paper, we address this general problem of the loss of social welfare that occurs due to uncoordinated behavior in networks and model it as a multiagent coordination problem. Specifically we study strategies to overcome selfish routing in traffic networks with multiple routes where a subset of vehicles are part of a social network that exchanges traffic related data. We investigate classic traffic flow paradoxes that are ubiquitous in various types of networks leading to severe congestion. We present a novel distributed traffic coordination algorithm that alleviates congestion by harnessing the real-time information available through the driver's online social network. We also propose a utility computation mechanism for route choice that generates near-optimal flows. Our extensive simulation results show that social network based multiagent traffic route coordination contributes to mitigate the effects of these paradoxes and significantly reduces congestion.