A traffic flow forecast supported system based multi-agent

The real time forecast of traffic flow is the function of intelligent transport systems (ITS), and no one of the forecasting methods dominates the others obviously in all conditions. An intelligent forecasting system built on many forecasting models is proposed and the emphases of the system is the expression of knowledge, design and implementation of the model base and the choosing the model which is implemented by multi-agent theory. The forecasting results of the system under simulation are compared with the results given by single model.