Meta-sezgisel algoritma tasarlamak planlama sorununu çözmek için paralel makinelerle ilişkisi olmayan görevlere erişim süresi sınırı endüstriyel otomasyon

This thesis investigates the optimization of scheduling tasks across unrelated parallel machines in industrial settings, focusing on minimizing the makespan to enhance operational efficiency. A genetic algorithm (GA), known for its robustness in solving complex optimization problems, is developed to address the scheduling challenges posed by machines with variable processing speeds and task dependencies. The study utilizes MATLAB to implement the GA, which intelligently explores potential scheduling sequences through sophisticated genetic operations such as crossover, mutation, and selection. An experimental setup involving four machines and twenty tasks is used to evaluate the algorithm's efficacy. The results demonstrate that the genetic algorithm significantly improves production efficiency by finding optimal task schedules. The research further discusses future enhancements, including the integration of additional constraints and the management of uncertainties in task processing times. This comprehensive study substantiates the capability of genetic algorithms to solve intricate scheduling problems, offering valuable insights and practical solutions applicable across diverse industries.