A Method of Traffic Dispersion for Accident Management

Traffic accidents usually induce non-recurrent congestions in the urban transportation network. In order to avoid the congestion deterioration, the paper presented a method to make traffic dispersion plans. The method considered and harmonized multi-objective, such as the change in total travel time, restrictions of traffic volumes on the links, and changes in traffic volumes on the links. Firstly, based on satisfactory control theory, the method introduced fuzzy goals and fuzzy constraints into a multi-objective function to obtain an optimal traffic dispersion plan. Secondly, a modified particle swarm optimization algorithm was raised to solve this complex multi-objective problem which has the characteristics of nonlinearity, multi-variable and multi-parameter. Finally, the method was operated on Matlab. The experimental results show that the method is feasibility and validity for accident management to disperse traffic and reduce the impact of the non-recurrent congestion.