Abstract: In order to improve the search efficiency of multi-objective evolutionary algorithms,a multi-objective Memetic algorithm based on simulated annealing is proposed.The method evaluates the individual fitness based on Pareto dominance relationship,applies simulated annealing to local search,and uses the crossover operator and a grid-density-based selection scheme to improve the convergence of the algorithm and to enhance the uniform distribution of solutions.Simulations on multi-objective flowshop scheduling problems show that the proposed algorithm can generate approximation sets closer to the Pareto front of the problem.