A Genetic Algorithm with Tabu Search for Multimodal and Multiobjective Function Optimization

The integration of genetic algorithms (GAs) and tabu search is one of traditional problems in function optimization in the GA literature. However, most proposed methods have utilized genetic algorithms to explore global candidates and tabu search to exploit local optimal points. Unlike such methods so far, this paper proposes a new algorithm to directly store individuals into multiple tabu lists during GA-iterations. The tabu lists inhibit similar solution candidates from being selected so often. The proposed algorithm is so simple but strong that we can solve both multimodal and multiobjective problems in the same manner. The paper describes the basic idea, algorithms, and experimental results.