Particles swarm optimization is a new evolutionary computation technique.Compared with genetic algorithm，the particles swarm optimization has the advantages of easy implement，few control parameters，and fast convergence to optimal solution in most cases.In order to obtain better multilayer coating structure，the particles swarm optimization is proposed for the design of optical multilayer coatings.To verify the feasibility of the method，three different design examples，i.e.，anti-reflection film，hign-reflection film and beam-splitting film，were demonstrated.In the examples，the sum of squared error of ideal reflectivity and actral reflectivity was used as the merit function of multilayer coating performance.The results indicate that the particles swarm optimization is an effective technique for the design of multilayer coatings.By means of particles swarm optimization，a better multilayer coating structure than genetic algorithm can be obtained under the same design condition.