Genetic Algorithm - Adaptive Crossover Based on Solution Distribution in Search Space

Genetic algorithms are probabilistic searching techniques that are able to provide results in complicated problems from various IT domains. The generated results depends directly by parameters of genetic algorithms operators (crossover and mutation probability, population size, number of generations, types of chromosomes, fitness function, etc.). In this paper it is proposed an adaptive type of crossover operator, that use both the fitness value of chromosomes involved and also their relative position in the search space relatively to other members of the population. We will demonstrate that, using this type of operator it can be obtain an improvement in terms of algorithm convergence to a global optimum solution.