Decision making in smart transportation networks

In this research project we study traffic networks from the point of view of game theory. The travellers on a traffic network can be seen as ‘players’, each of which is trying to minimize the needed time to travel from its origin to its destination by making use of the ‘resources’ available, namely, the roads. As individuals make choices about which routes to take, popular roads become digested, and travel time increases accordingly. Eventually the network may settle in equilibrium, where the amount of traffic per road is constant, and all players are making optimal choices for the given network state. Such a constant state is called a ‘Wardrop equilibirum’ and this research project focuses on the properties of these equilibria under different assumptions. This set-up is known in the literature as a ‘routing game’ or a ‘congestion game’. Besides traffic networks there are other applications of this field of research, one instance of which is the design and control of communication networks.

Initially we will be looking at the inclusion of risk-averseness in the considerations made by the different players in a routing game. In addition to minimizing the costs (i.e. travel time), players may also want to limit the risks that they are taking by choosing a certain route. Various measures of risk may be considered and the existence and properties of the associated Wardrop equilibria effects studied.