Multi-Objective Sequential Route Optimization Applied In Large-Scale Road Networks With Traffic Congestion

The complexity of real-world problems requires, in most cases, optimized solutions considering multiple objectives. For this reason, the multi-objective optimization has been increasingly used to treat this kind of problems. In this work, an approach is proposed to deal with multi-objective routes generation considering multiple metrics and traffic congestion estimates. The experiments include vehicles that intend to perform routes with multiple stops in large-scale road networks. The OpenStreetMap data was used to create the road network that contains all information needed. Four scenarios were simulated with different levels of traffic congestion. After this, the obtained results were compared with the best solutions computed by Dijkstra's Algorithm. The proposed approach has obtained good computational performance and shown efficiency, offer good trade-offs, highlighting the best results for scenarios with higher traffic congestion levels.