A study on real-coded genetic algorithm for process optimization using ranking selection, direction-based crossover and dynamic mutation

In this paper, a novel and efficient real-coded genetic algorithm (RCGA) for process optimization is developed. The proposed RCGA is equipped with Ranking Selection (RS), Direction-Based Crossover (DBX) and Dynamic Random Mutation (DRM) operators. The RS operator is used to eliminate the bad solutions and reproduce good solutions, making the whole population to achieve a better average fitness. The DBX operator uses relative fitness information to direct the crossover toward a direction that significantly improves the objective fitness. The DRM operator prevents the premature convergence of RCGA and at the same time increases the precision of the searched solution. The effectiveness and application of the proposed RCGA are demonstrated through a variety of single objective optimization benchmark problems. For comparative study, other existing RCGAs with different evolution operators are also performed to the same problem set. Extensive experiment results reveal that the proposed RCGA provides a significantly faster convergence speed and much better search performance than comparative methods.