Cultural-Based Genetic Algorithm: Design and Real World Applications

Due to their excellent performance in solving combinatorial optimization problems, metaheuristics algorithms such as genetic algorithms GA (Sareni and Krahenbuhl, 1998; Karr and Freeman, 1999; and Chambers, 1995), simulated annealing SA (Reznik, 1997 and Gill et al., 1981) and tabu search TS make up another class of search methods that has been adopted to efficiently solve dynamic optimization problem. Most of these methods are confined to the population space and in addition the solutions of nonlinear problems become quite difficult especially when they are heavily constrained. They do not make full use of the historical information and lack prediction about the search space. Besides the knowledge that individuals inherited "genetic code" from their ancestors, there is another component called culture. In this paper, a novel culture-based GA algorithm is proposed and is tested against multidimensional and highly nonlinear real world applications.