Evolutionary algorithm with a directional local search for multiobjective optimization in combinatorial problems
AbstractThis abstract summarizes the results reported in the paper . In this paper a new method of performing a local search in multiobjective optimization problems is proposed. The proposed method uses a solution acceptance criterion based on aggregation of the objectives using adaptively adjusted weight vectors. A weight vector for performing the search starting from an initial solution is determined using directions in which objective improvements have been achieved in the vicinity of the initial solution. In the paper the proposed method is tested on 2-, 3- and 4-objective instances of the Travelling Salesman Problem (TSP) and the Quadratic Assignment Problem (QAP). In the experiments the proposed method outperformed two other local search methods. The proposed method focuses on solution acceptance criterion and thus can be combined with various methods of solution neighbourhood construction in the local search as well as various global search algorithms
|Publication size in sheets||0.3|
|Book||Bosman Peter (eds.): Proceedings of the Genetic and Evolutionary Computation Conferenc (GECCO'17 Companion), 2017, Association for Computing Machinery, ISBN 978-1-4503-4939-0, 1893 p.|
|Publication indicators||= 2|
* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.