Dynamic Fuzzy Logic Control of Genetic Algorithm Probabilities

Genetic Algorithms are traditionally used to solve combinatorial optimization problems. The implementation of Genetic Algorithms involves of using genetic operators (crossover, mutation, selection, etc.). Meanwhile, parameters (such as population size, probabilities of crossover and mutation) of Genetic Algorithm need to be chosen or tuned. In this paper, we propose a hybrid Fuzzy-Genetic Algorithm (FLGA) approach to solve the multiprocessor scheduling problem. Based on traditional Genetic Algorithms, a fuzzy logic controller is added to tune parameters dynamically which potentially can improve the overall performance. In detail, the probabilities of crossover and mutation is tuned by a fuzzy logic controller based on fuzzy rules. Compared to the Standard Genetic Algorithm (SGA), the results of experiments clearly show that the FLGA method performs significantly better.