Learning from user's behaviour of some well-known congested traffic networks
This addresses the traffic assignment problem for urban planning or transportation systems, but appears incremental as it applies existing methods to a known bottleneck.
The paper tackles the problem of predicting user behavior in congested traffic networks under equilibrium conditions, proposing a two-stage machine learning approach that combines a neural network with a fixed point algorithm and evaluates it on classical networks.
We consider the problem of predicting users' behavior of a congested traffic network under an equilibrium condition, the traffic assignment problem. We propose a two-stage machine learning approach which couples a neural network with a fixed point algorithm, and we evaluate its performance along several classical congested traffic networks.