An Application of Particle Swarm Optimization Algorithm for TSP Problem

Particle Swarm Optimization (PSO) is a population-based optimization algorithm inspired by the collective foraging behavior of bird flocks. When applied to the Traveling Salesman Problem (TSP), PSO simulates particles searching through the solution space, continuously adjusting their positions (representing route solutions) to find the optimal path. In your graduation project, PSO is implemented to solve a 51-city TSP problem. The objective of TSP is to find the shortest possible route that visits each given city exactly once and returns to the starting point. PSO's advantages lie in its simple implementation, fast convergence speed, and suitability for discrete optimization problems. The core PSO mechanism guides particle updates through individual best (pBest) and global best (gBest) positions. For TSP implementation, particle positions can be encoded as city permutation sequences, while the fitness function typically calculates total path distance. To satisfy TSP constraints (each city visited exactly once), permutation operations or discrete PSO variants are commonly employed in the code structure. Furthermore, you plan to compare PSO results with Genetic Algorithm (GA) performance. GA is another popular optimization method that evolves path solutions through selection, crossover, and mutation operations. While GA typically handles complex combinatorial optimization more effectively, it may exhibit slower convergence compared to PSO. By testing both algorithms on identical city datasets, you can analyze their relative strengths in convergence speed, solution quality, and computational efficiency. Your experimental framework can be extended to examine how different city scales (e.g., 30, 50, or 100 cities) affect algorithm performance, or to optimize convergence behavior by adjusting PSO parameters like inertia weight and learning factors. This research ultimately provides valuable insights into optimization methodology selection for TSP problem-solving.