A Genetic Algorithm Balancing Exploration and Exploitation for the Travelling Salesman Problem

This paper presents an investigation to the genetic algorithms (GAs) that have been successfully applied to solve many combinatorial problems. To the general problem of premature convergence to local rather than global optima due to lack of explorative capabilities of the algorithm in the GA research field, this paper proposes a novel approach improving the explorative capabilities and the exploitation effects. The proposed algorithm is studied to balance the exploration to a great diversity of tours and the exploitation of excellent individuals, called Bee-GA. And empirical tests using the traveling salesman problem (TSP) as the case application in order to quantify its performance have shown that the Bee-GA performs highly competitive in terms of solution quality.