Research on county traffic optimization model based on spatio-temporal big data analysis

This investigation endeavors to address the challenging issue of forecasting and enhancing traffic flow through the development of a county-level traffic optimization framework grounded in spatio-temporal big data analytics. Toward this objective, the present work introduces a composite approach integrating a convolutional long short-term memory network with a particle swarm optimization technique. Initially, ConvLSTM is employed to discern the intricate patterns within spatio-temporal traffic datasets, thereby precisely grasping the fluctuations and interdependencies of traffic volumes over space and time. Building upon these insights, the PSO method refines the parameters governing traffic signal operations and route planning to ascertain the attainment of an optimal solution. This framework not only enhances the precision of traffic flow projections but also facilitates the holistic improvement of traffic networks. Consequently, it furnishes a novel and viable strategy for managing county traffic, aiding in the alleviation of congestion and boosting traffic fluidity.