Q. Chen, J. Zhang

Pages: 69-82

Abstract
In the basic sections of highways, the traffic flow is relatively constant, and the traffic capacity and the service level can generally meet the designed requirements. However, in the confluence zones, the traffic confluence among the vehicles will often lead to traffic congestion as the traffic volume increases, especially at the position where the vehicle from a tollbooth enters the main lane. Therefore, the paper uses the models of Queuing Theory and Cellular Automata and employs the MATLAB software to model the traffic situation near a tollbooth. This paper found the optimal number of tollbooths B corresponding to the number of lanes of travel L. Using the Alexander Computer Simulation Technology, the paper establishes the merging model of traffic flow in the confluence zone. On the basis of the existing toll plaza, the paper designed a gourd-shaped lane consolidation zone. Then, the Markov Model is used to analyse the designed merging method. This model proves the superiority of this design in parameters such as throughput, construction costs, traffic velocity, delay time, traffic density and more. Finally, the paper uses the Queuing Theory model to study the relationship between the traffic flow and the ratio of different toll types. For different traffic flows, the proportion between the number of MTCs and the number of ETCs is different. The model provides a reasonable suggestion for the construction of highway toll plazas.
Keywords: queuing theory; cellular automata; toll plaza; Markov model