Flexible job-shop scheduling is a very important branch in both fields of production management and combinatorial optimization. A hybrid particle-swarm optimization algorithm is proposed to study the mutli-objective flexible job-shop scheduling problem based on Pareto-dominance. First, particles are represented based on job operation and machine assignment, and are updated directly in the discrete domain. Then, a multi-objective local search strategy including Baldwinian learning mechanism and simulated annealing technology is introduced to balance global exploration and local exploitation. Third, Pareto-dominance is applied to compare different solutions, and an external archive is employed to hold and update the obtained non-dominated solutions. Finally, the proposed algorithm is simulated on numerical classical benchmark examples and compared with existing methods. It is shown that the proposed method achieves better performance in both convergence and diversity.