<p>Assembly line balancing is one critical problem that is common in the manufacturing and sewing of garments, as it determines the efficiency and productivity of an organization. The aim of this research is to develop a bi-objective genetic algorithm to optimize task assignments with the objectives of minimizing task imbalance and minimizing movement of semi-finished products. The algorithm applies order crossover and uniform mutation to increase population diversity and exploit the solution space effectively. A case study within the garment industry illustrates the comparison of the methods with the existing task assignment strategy. Two approaches are used: (i) the Epsilon-constraint method, and (ii) a hybrid model that combines genetic algorithms with linear programming in a two-loop structure. By applying the two methods, significant improvements were achieved in task distribution, and production flow, while unnecessary product movement was minimized. The solution is well-defined for practicality, scalability, and adaptability for improving other production assembly lines in different industries facing similar challenges, which provide the manufacturers with the means to enhance productivity and competitiveness in a fast-changing marketplace.</p>