Research On Merging Area’s Capacity Of Beijing Urban Expressway Interchange

Interchange is an important node of urban expressway network. Enhancement of thecapacity of the interchange is the key to improve the capacity of the whole urbanexpressway systems. In recent years, many cities in China have developed their urbanexpressway systems. Though remarkable progress has been shown, there is still muchleft to be improved since the insufficient theoretical research in this field, especiallyresearch on the traffic capacity of the each component as well as the whole interchangesystem. Potential insecurity lies in the way of future expressway system planning, designand selection evaluation. Yet this thesis chose the capacity of merging areas ofexpressway interchange as entry and focus, due to the fact that the confusion often takenby the merging behavior of ramp vehicles in merging area, the speed of vehiclesbecomes lower and the travel delay increases, thus also labeled as the bottleneck of theexpressway. The research was inspired and supported by Studies on the Traffic Capacityof Beijing urban expressway interchanges.In this context, the research in this dissertation develops models on the capacity ofmerging areas of Beijing urban expressway interchange. Frist, data was collected tocategorize the types of merging areas. The result showed that parallel type accelerationlanes are the major type on the Third and Fourth Ring Road, accounting for around80percent of all types. Thus, this research took parallel type acceleration lanes as the studyobject. Second, traffic survey was taken of typical merging areas of the Third Ring Roadusing VIP-T. The data provides a solid basis for the mathematical model feasibility ofthis study. Then, GA-BP traffic capacity model was developed, as well as the simulationmethod using VISSIM software. Subsequently, a case study on Lize Bridge wasconducted on the basis of the proposed models. It is demonstrated that, comparing withthe results calculated from the two models, the absolute error is80pcu/h while theGA-BP traffic capacity models performs better. Finally, it provides recommendations infuture studies.