With the surge in vehicle numbers and urbanization, modern cities face escalating traffic management challenges. This research presents AI-driven solutions to enhance green transportation by leveraging predictive analytics and deep learning. A comprehensive framework that combines Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to capture spatial and temporal traffic patterns is proposed. By employing Particle Swarm Optimization (PSO) and Bayesian Optimization, optimal performance is theorized. Additionally, our approach integrates Deep Reinforcement Learning (DRL) to facilitate real-time traffic management, dynamically adjusting to varying conditions to reduce congestion and promote efficient transportation. This research offers innovative strategies for sustainable urban mobility, emphasizing the potential of AI in transforming traffic systems for greener cities. The focus on the intersection of AI and urban mobility underlines how key technologies can help resolve major contemporary challenges in urban transportation.