A Hybrid Cooperative Guided Local Search for Solving the Traveling Salesman Problem

In this paper, a hybrid cooperative Guided Local Search (GLS) algorithm for solving the Travelling Salesman Problem (TSP) is proposed. The idea of the algorithm is to run a number of agents of GLS and after a number of local search calls, the agents exchange information about their search utilizing the crossover operator to make their further search rational. The partially mapped crossover (PMX) is applied based on the penalty’s values of features to preserve the good components of the best solutions found by the agents during the search. The preliminary experimental results demonstrate that the proposed algorithm is promising and outperforms parallel GLS without collaboration.