An Improve Artificial Immune Algorithm for Solving The Travelling Salesman Problem

This paper proposes a new technique, which is an adapted of Artificial Immune Algorithm (AIA) for solving the well-known Travelling Salesman Problem (TSP). The aim of our algorithm is to achieve better optimal solution than the original version of the AIA. The proposed algorithm introduces two new features to the standard AIA algorithm. (1) the Double Split Method is introduced to help increase the solution diversity in the clonal selection stage; (2) the addition of K-means clustering algorithm to group the same solution into one group. The proposed algorithm is tested with 10 well-known benchmark datasets. The result demonstrates that the proposed algorithm is outperforming than the comparison algorithms in term of the average percentage deviation from the lower bound.