The current industry faces strong demands for high quality, low-cost personalized products with shorter deadlines, driven by international market expansion and globalization. Manufacturing systems have evolved significantly to incorporate greater flexibility, enabling rapid adaptation to changing production needs, customization of products, and efficient utilization of resources. Flexible manufacturing systems are adopted to enhance productivity and reduce production times. The Flexible Job Shop Scheduling Problem (FJSP) plays a pivotal role in modern manufacturing systems by supporting adaptability, efficiency, customization, competitiveness, and the integration of advanced technologies across various industries. The FJSP deals with the challenge of optimizing job allocation and sequencing in a manufacturing system where machines can perform various types of operations, aiming to improve productivity and adaptability in dynamic production environments. In this paper, we propose a bi-level genetic algorithm-based tabu search, called Bi-GenTab, to solve the FJSP in order to minimize the maximum completion time, the critical workload and the total workload. To evaluate the performance of our approach, we carry out experiments on Kacem and Brandimarte benchmark instances. The results of the experiments show the efficiency of the Bi-GenTab in comparison with other known methods in the literature.