Eco-Routing in Urban Networks for Smart Cities: A Hybrid PSO-GA Approach for Low-Carbon Traffic Paths

Sustainable urban mobility demands routing strategies that curb greenhouse-gas emissions, not just travel time or distance. We present a hybrid metaheuristic that fuses Particle Swarm Optimization (PSO) with a Genetic Algorithm (GA) to discover low-carbon routes in real-world street networks extracted from OpenStreetMap. Each road segment is assigned a carbon cost proportional to its length and an emission factor; the PSO phase rapidly guides the search toward promising solutions, while GA operators diversify and refine the path population. Applied to the dense and heterogeneous networks of Mumbai, Istanbul, Rome, and Bangalore, the PSO-GA method cuts cumulative CO2 cost by up to 47% relative to the classic Dijkstra shortest-path baseline, yet maintains practical travel lengths. Experiments show fast convergence, near-linear runtime scaling, and straightforward parameter tuning—key attributes for smart-city deployment. The framework is extensible to multi-objective routing and can ingest live traffic or sensor data, making it a viable core for eco-aware navigation services that align urban transport with climate goals.