This study studies the increasing complexity of modern manufacturing scheduling, where efficiency, quality, and sustainability must be jointly optimized under flexible machine and operator constraints. Integrating real-time feedback from digital twins into optimization frameworks has emerged as a powerful approach, enabling adaptive and data-informed decision-making. By combining exact methods and metaheuristics, such frameworks can navigate the multi-objective landscape of contemporary
production systems effectively. Looking forward, the adoption of surrogate models offers a promising alternative to further enhance performance. By approximating expensive simulations or high-fidelity digital twin responses, surrogate models can significantly reduce computational costs while maintaining solution accuracy.