Optimizing Urban Mobility through Dynamic Toll Pricing: A Predictive Traffic Modelling Approach for Revenue Generation, Congestion Management, and Environmental Sustainability

The combination of city development with increasing vehicle population in metropolitan areas generates three principal problems linked to traffic congestion and poor toll structures and environmental deterioration. The Hybrid Dynamic Toll Pricing Framework developed by scientists uses machine learning models and geo-clustering methods with reinforcement learning methodology to enhance urban mobility. The system uses present-time traffic information and projected traffic data and congestion metrics to adjust toll rates that reach maximum efficiency together with revenue generation and environmental sustainability. Toll boundaries definition depends on DBSCAN-based geo-clustering methods but the system integrates this information with traffic prediction through LSTM, ARIMA and Prophet modeling and Q-learning regulates automatic toll pricing adjustments. The simulation results evaluate the system's dual functionality by assessing both its impact on decreasing traffic congestion and its revenue generation potential as well as environmental benefits. The implementation of dynamic toll pricing achieves optimal results in reducing congestion and revenue increase while reducing environmental emissions due to decreased fuel consumption and waiting durations according to simulation analysis. Different subsidy plans presented in the research examine methods to keep transportation expenses fair for low-income bracket customers. The adaptive toll system serves urban transit by developing an efficient sustainable transportation solution with potential applications for future infrastructure planning and development of transportation systems.