Improving genetic algorithms’ performance by local search for continuous function optimization Cos ßkun Hamzac ¸ebi Z. Karaelmas University, Department of Informatics, 67100 Zonguldak, Turkey Abstract The genetic algorithms (GAs) can be used as a global optimization tool for continuous and discrete functions problems. However, a simple GA may suffer from slow convergence, and instability of results. GAs’ problem solution power can be increased by local searching. In this study a new local random search algorithm based on GAs is suggested in order to reach a quick and closer result to the optimum solution. Ó 2007 Elsevier Inc. All rights reserved. Keywords: Genetic algorithms; Local search; Random search; Function minimization 1. Introduction A typical optimization problem based on a continuous function f(x)– f(x) is function of x ¼fx 1 ; x 2 ; ... ; x n g; x 2 S ; S is the search space and S  R n -deals with to find the value of x best in the search space S which make f(x best ) is optimal for all x values. Optimization techniques can be classified into two classes: deterministic, and stochastic. Deterministic methods require the existence of derivatives and the continuity. If f(x) is continuous and differentiable, the optimum value can be found of the point where @f @x ¼ 0. However, when the function is not differentiable, it could prove more advantageous to utilize stochastic methods instead of deterministic ones. Random search, hill-climbing, simulated annealing, Tabu search, ant colony and GAs are stochastic meth- ods, used in function optimization. These methods can be used in both discrete and continuous function opti- mization problems. In literature different stochastic optimization algorithms were used for minimization or maximization of a function. For example, Li and Rhinehart [1] have proposed the heuristic random optimi- zation (HRO), Kwon et al. [2] have suggested successive zooming genetic algorithms (SZGA), Ji and Tang [3] have proposed memory Tabu search (MTS), Hamzac ¸ebi and Kutay [4,5] have suggested adaptive random search technique (ARSET), and dynamic random search technique (DRASET), Toksari [6,7] has proposed ant colony optimization (ACO), and a modified version of ACO named as MACO. 0096-3003/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.amc.2007.05.068 E-mail address: coskunh@karaelmas.edu.tr Available online at www.sciencedirect.com Applied Mathematics and Computation 196 (2008) 309–317 www.elsevier.com/locate/amc