Dynamic Routing in High-Density Urban Networks: A Hybrid Swarm Intelligence Approach to Last-Mile Logistics

The exponential growth of global e-commerce has heavily strained urban logistics, making last-mile delivery the most expensive and time-consuming segment of the supply chain. In dense urban environments, traditional routing algorithms often fail to adapt to dynamic traffic congestion, strict customer time windows, and capacity constraints. This paper introduces a novel computational model, the Hybrid Swarm Intelligence Algorithm (HSIA), which synergizes Ant Colony Optimization (ACO) with Particle Swarm Optimization (PSO) to solve the dynamic Capacitated Vehicle Routing Problem with Time Windows (CVRPTW). Designed for presentation at ICCMETS 2026, this study leverages computational modeling to optimize fleet deployment. In our framework, ACO is utilized for discrete route construction based on historical traffic data, while PSO continuously fine-tunes the pheromone trails and dynamically re-routes vehicles in response to real-time traffic anomalies. Simulated experiments on modified Solomon benchmark instances—adjusted to mimic the high-density road networks of major metropolises—demonstrate the superiority of the proposed framework. The HSIA achieved an 18.4% reduction in total travel distance and a 14.2% decrease in fuel consumption compared to standard Genetic Algorithms and standalone ACO models, while maintaining a 98% on-time delivery rate. These findings offer scalable technology solutions for modern logistics enterprises aiming to minimize operational costs and carbon footprints in smart cities.