Analysis Of The Congestion Drift And Research On Induction Mechanism In The Traffic Routing

The traffic congestion problem and the safety problem are important problems which involve the livelihood of the people. At present, the researches on the planning and the control of the urban traffic have become more and more mature, but the two measures still can't solve the traffic congestion problem and the safety problem very well. For example, regarding to the complex traffic system, the two measures can't solve the traffic congestion problem and the safety problem well only using road planning or vehicle control methods. Therefore, the intelligent traffic system, such as the vehicle navigation system, is adopted to control and manage the road network to increase utilization efficiency of the traffic system. With the development of intelligent traffic system, the traffic guidance mechanism is gradually applied to the traffic system. Modern communication technology and computer technology have been used to induce the travelers to choose the best route, induce the travelers not to choose the crowded route to travel, and reduce traffic congestion phenomenon. By providing the better route suggestions, it reduces not only the time which the travelers required, but also the burden of the traffic system. Therefore, it is important to induce the travelers to choose the route effectively by the traffic guidance system.At present the researches on the traffic guidance mechanism are popular at home and abroad, but most of the researches all apply different algorithms to provide an unobstructed road or provide the information about the crowded road, which may make most of the travelers change the way to the induced road, or run clear off the crowded road which may make the alternative road generate congestion and result in the congestion drift. This is due to the adverse induced result which not only brings the travelers and the whole traffic system inconvenience, but also makes the trust level of the guidance mechanisms decrease.In this paper, the idea of game theory is introduced to solve the traffic guidance problems. Traffic system is dynamic, in the dynamic traffic system, there are the game behaviors between traffic managers and travelers and among travelers. The purpose of travelers is to make their own travel utility maximize, that is. the travelers will not regret making the choice, and the traffic system reaches to the Nash equilibrium at this moment. The purpose of traffic management is to make the travel time shortest and the traffic system most effective. The Game Theory is introduced to resolve the conflict between traffic managers and travelers. We verify the gap between the user optimization and the system optimization and find that the induction is necessary through mathematical derivation and numerical calculation. According to the characteristics of traffic system, we establish a simple path model and based on this model provide a new induction mechanism. In this mechanism, the managers formulate the inducing strategies based on the number of travelers, the conditions of the road and the other information. According to the principle of system optimization, coefficients are modified and then the modified information is released to the travelers to induce them to select the path. We control the system from the macro, calculate the total time required for the travelers under the inducement and find that it is much shorter than the time required for no inducement. Besides it is close to the optimal time. When the capacity of the path in the traffic system is close to saturation, we induce travelers to choose other paths, but not only provide one or two alternate paths. Using this method, congestion drift caused by few alternative paths can be avoided and the traffic system operating conditions can be improved, the total time required for the whole traffic system can be reduced and congestion drift phenomenon before the traffic system reaching the saturation capacity can be avoided. This induction mechanism improves the utilization of the traffic network, and makes the travelers trust the induced traffic system greatly. We verify the effectiveness of this induction mechanism through simulation experiment in this paper. We extend the simulation experiments to a complex road network, and verify that the induction mechanism is effective in this complex road network. The study on this induction mechanism plays an important role for the development and applications of traffic guidance system undoubtedly.