Collaborative optimization strategy for urban highways and roads based on electronic toll collection lane regulation

To address the issue of coordinated optimization between urban highways and city roads, and to enhance traffic flow efficiency while alleviating congestion, the research proposes a lane control method based on an electronic toll collection system. The study utilizes long short-term memory network algorithms to achieve accurate forecasting of traffic flow and dynamic adjustment of lane opening strategies. Concurrently, an in-depth examination is carried out on the lane control mechanism embedded within the electronic toll collection system, alongside the development of a collaborative optimization model tailored for urban highways and city roads. The LSTM algorithm is utilized for traffic flow prediction. The experimental results indicate that the proposed model outperforms other models in terms of processing time, resource utilization, training time, regression loss, and convergence speed. On the METR-LA dataset, the proposed model has a processing time of 25.34 s, a resource utilization rate of 78.4%, a training time of 15.5 h, a regression loss of 0.0237, and a convergence speed of 120 iterations. Furthermore, the proposed model demonstrates strong performance in various prediction scenarios, such as short-term prediction, where the model delay is 9.67 s, the average traffic flow is 1600 vehicles/h, and the prediction stability reaches 90.4%. The significance of this research lies in the fact that the proposed model provides an effective tool for urban traffic management, capable of optimizing traffic flow in real-time, reducing congestion, and improving road utilization efficiency.