Further Enhancement of Genetic Algorithm Using Multiple Crossover and Mutation Operators

This research focuses on further enhancing the performance of genetic algorithms through the utilization of multiple crossover and mutation operators. The primary objectives of this study are to improve the effectiveness of genetic algorithms by employing diverse crossover and mutation operators, develop a novel crossover operator to overcome limitations of existing operators, and propose innovative crossover and mutation operators to address local optima issue. The methodology employed in this research centers around the Nurse Scheduling Problem, which involves creating an optimal schedule for a group of nurses considering various constraints. Genetic algorithms are used to solve this problem, with binary encoding representing the nurse schedule. To enhance the algorithm, three crossover algorithms (Poor and Rich Optimization, Golden Search Algorithm, and Prairie Dog Optimization Algorithm) and three mutation algorithms (Chaotic Vortex Search Algorithm, Capuchin Search Algorithm, and Chaos Cloud Quantum Bat Hybrid Optimization) are proposed. The research findings indicate that the combination of Prairie Dog and Capuchin Search algorithms outperforms other combinations, resulting in a significantly lower number of shift violations (only 7 violations). Overall, this research contributes to the field of genetic algorithms by presenting novel crossover and mutation operators that enhance their performance, particularly in tackling the Nurse Scheduling Problem.