A modelling of route choice behaviour in transportation networks : an approach from reinforcement learning.

We present a simple, but very powerful traffic equilibrium calculation method. The basic idea of the method is motivated from reinforcement learning with profit sharing. In our model, individual driver is regarded as heterogeneous entity, being assumed to form his or her own value for each route through driving experiences and communications to the environment. Our method realizes a disaggregate user equilibrium on a congested network so that it is useful to analyse the interrelationships between each driver's characteristic and the resultant network equilibrium. Moreover, this method not only covers from stochastic user equilibrium to deterministic user equilibrium, but it is also applicable to a network with asymmetric cost functions or with discontinuous cost functions. For the covering abstract see ITRD E128680.