Parameters Evaluation of Ant Colony Algorithm based on TSP

2016 Revista Técnica de la Facultad de Ingeniería Universidad del Zulia  
Ant colony algorithm is a kind of new ecological system algorithm used to solve complicated combinatorial optimization problems, and it is also a kind of heuristic evolutionary algorithm based on population, which belongs to a kind of random search algorithm, and is used to solve Traveling Salesman Problem( TSP). As you can see in the mathematical model of ant colony algorithm, the parameter space of the algorithm is large and the reasonable parameters' setting helps to improve the convergence
more » ... peed and global optimization ability of the algorithm. Influencing the parameters of the algorithm performance mainly have three aspects: pheromone heuristic factor α, expected heuristic factor β and pheromone evaporation coefficients ρ. This paper firstly analyzes how each parameter affects search capability and convergence speed of ant colony algorithm, and then chooses TSPLIB. eil51 databasebased on TSP problem as an example to validate the relationship between the algorithm performance and parameters ,lastly which are given the optimal selection range of each parameter and the best configuration .
doi:10.21311/ fatcat:6booyaq5mbaubcloa3n2jmpslu