EV charging station selection and routing flask application with ACO and NSGA-II including photovoltaic energy constraints

This paper presents an intelligent electric vehicle routing and charging optimization framework designed to minimize travel distance, energy consumption, and charging costs while maximizing the integration of photovoltaic energy. The proposed hybrid model combines Ant Colony Optimization for efficient route planning with the Non-dominated Sorting Genetic Algorithm II for multi-objective optimization under realistic conditions, including State of Charge, vehicle speed, charger occupancy, dynamic pricing, and renewable energy availability. Implemented as a Flask application with real-time communication via MQTT, the framework was validated across five realistic driving scenarios in Morocco, encompassing urban, suburban, and highway routes. Results demonstrate that the proposed algorithm significantly outperforms traditional A*, Ant Colony Optimization, and Non-dominated Sorting Genetic Algorithm II approaches, reducing travel distance by 7–10% and energy consumption by 10–15% compared to A*, and achieving up to 15–20% faster travel times than Non-dominated Sorting Genetic Algorithm II. In scenarios with photovoltaic availability, it lowers charging costs by 30–40%, increases renewable energy utilization remains high (≈70–98%) achieving ≈3–4 kg CO₂ reduction per trip relative to grid-only charging. Overall, the hybrid PV-aware optimization framework surpasses all benchmark algorithms by jointly enhancing efficiency, economic performance, and environmental sustainability, establishing a scalable foundation for next-generation electric vehicle routing and charging management in smart cities.