Research on the Application of Deep Reinforcement Learning based on Big Data in the Optimization of Intelligent Transportation Systems

In this study, we propose an Intelligent Transportation Optimization Network (ITON) that integrates deep reinforcement learning (DRL), big data analysis, and dynamic adjustment mechanism to improve traffic flow management efficiency, reduce congestion, and reduce carbon emissions. By leveraging large-scale historical and real-time traffic data, we first pre-trained with an improved Double Q network (Double DQN) to capture basic traffic patterns. Subsequently, the model is fine-tuned based on real-time data to ensure that it is adaptable and robust. In addition, a dynamic adjustment module based on attention mechanism is designed to continuously optimize the decision-making process through real-time feedback, which significantly improves the response speed and accuracy of the model in complex environments. The experimental results showed that ITON reduced the average travel time by 34.2%, the congestion index by 33.8%, and the carbon emissions by 42.9%. This method not only provides new ideas for the optimization of intelligent transportation systems, but also provides valuable references for the application of other fields such as industrial automation and energy management.