This paper is devoted to the study of a multi-criteria scheduling prob- lem on unrelated processors with machines’ learning effect, with the goal of minimizing makespan, machine cost and maximal flow-time simultaneously, which is an NP-hard problem. An improved particle swarm optimization algorithm equipped with the over- loaded operators, as well as a procedure of Levy flight, is proposed to generate the Pareto-optimal solutions. The experimental results show that the Levy flight strategy can effectively improve the performance of the algorithm, which can generate more non-dominated solutions, and slightly reduce the execution time of the process.