On-line distributed prediction of traffic flow in a large-scale road network

For on-line traffic control at traffic control centers there is a need for fast computations of predictions of traffic flow over a short prediction horizon, say 30 min, to evaluate the impact of different scenarios for the purpose of on-line scenario selection. A novel approach is presented to predict the traffic flow in a large-scale traffic network in an asynchronous, parallel, and distributed way at two or more subnetworks combined with a consistency check at the network level within a reasonable-small computation time.