This project compares genetic algorithms (GA), particle swarm optimization (PSO), ant colony optimization (ACO), and Transformer-based algorithms, and discusses which algorithm is more suitable for solving airport problems.
Based on experiments, we believe that simple scenarios with less than 50 nodes are suitable for GA algorithms, dynamically changing environments are suitable for ACO algorithms, and complex constraints with 200+ nodes are suitable for Transformer-based algorithms. Different algorithms can be selected for TSP problems according to different situations.