Optimizing smart transportation systems using AI techniques such as reinforcement learning (RL) involves tackling complex, multi-objective combinatorial problems. This research aims to develop a robust methodology for applying and evaluating RL and other AI methods to enhance traffic management, route planning, and resource allocation. A critical aspect of this work is the integration of explainable AI techniques to ensure that the decision-making process is transparent and interpretable, providing insights into how these AI models optimize system-level mobility in real-world transportation networks.