Comparisons between the Hybrid Taguchi-Genetic Algorithm and Genetic Algorithm


Koun-Tem Sun, Ching-Ling Lin, Hsin-Te Chan, Hong-Ming Kang, Man-Ting Ku


Several nations across the globe are actively pursuing the exploration of space through space observatories, robotic systems, and humans. These efforts are enabled by the development of needed transformative technologies and methodologies. The complexity of these efforts is underscored by the need to use planetary resources and reuse resources taken from Earth to ensure the safety and cost effectiveness of future exploration missions. With revolutionary logistics methodologies, these efforts will enable sustainable and affordable space exploration. The importance of new technologies for energizing the economic engines of many nations is making it essential to transfer space technologies to the commercial sector. The U.S., through NASA, has impleHybrid Taguchi genetic algorithm can be used to solve the global continuous optimization problems. Aside from the global search capability of traditional genetic algorithm, it further combines Taguchi experimental method to explore the optimal feasibility of the offspring. Taguchi method is inserted between the crossover and mutation operations of the traditional genetic algorithm. Hybrid Taguchi genetic algorithm also seems to outperform the traditional genetic method in obtaining the optional or near optimal solutions because of its fast convergence ability and robustness. Although the hybrid Taguchi genetic algorithm is more powerful than the traditional genetic one in the optimization of global continuous function, yet it still needs further investigation to conclude if it also offers better solution than the latter to the optimization of global discrete function. Therefore, this study tries to compare the two algorithms in each individual’s performance in the optimization of global discrete function. It aims to figure out whether the hybrid Taguchi genetic algorithm is better than traditional genetic algorithm or not.


Hybrid Taguchi-Genetic algorithm, Genetic algorithm, Optimization problems.