The rapid advancement of autonomous vehicle (AV) technologies necessitates the advancement of intelligent traffic management systems that can ensure efficient, safe, and adaptive navigation. This research presents a novel framework for Intelligent Traffic Routing Systems (ITRS) for autonomous vehicles using a deep learning-based optimization algorithm. The proposed system integrates predictive deep learning models with evolutionary optimization techniques such as Genetic Algorithms (GA) to dynamically analyze real-time traffic data and identify optimal routing paths. The deep learning component is trained on historical and live traffic datasets to forecast con-gestion patterns, travel times, and road conditions. These predictions serve as input to the optimization algorithm, which evolves route solutions through iterative selection, crossover, and mutation processes, optimizing for factors like minimal travel time, reduced congestion, and safety. The integration of machine learning with bio-inspired optimization enables autonomous vehicles to make proactive and intelligent routing decisions in highly dynamic traffic environments. Simulation findings reveal considerable enhancements in routing proficiency and adaptability contrasted to classical routing frameworks, highlighting the potential of this approach in next-generation smart transportation systems. © 2025 Elsevier B.V., All rights reserved.