Abstract: The society’s demand for individualized, small batch and customized products can be satisfied by the flexible job shop system, this workshop system has numerous equipment, complex process paths, and various failure frequencies. However, the single machine preventive maintenance approach is now mainly employed to avoid equipment breakdowns, which is certain to raise the number of maintenance, maintenance costs, and have the negative impact on production operations. To address the problems caused by the traditional single machine preventive maintenance, the group preventive maintenance approach is proposed to apply in the flexible job shop system, and the joint mathematic model of group preventive maintenance and multi-criterion flexible job shop scheduling is established. In order to overcome the problem of insufficient local search ability of traditional algorithms, a new multi-criterion evolutionary algorithm is designed to solve the multi-objective flexible job shop scheduling problem, and to show how to apply the group preventive maintenance strategy in the flexible job shop system. The experimental results show that the designed multi-evolutionary algorithm can obtain more optimal solutions, has faster convergence speed, and can converge to better optimal solutions; compared with the single preventive maintenance method, the group preventive maintenance has fewer maintenance times, lower maintenance costs, and less impact on production activities, the example result show that the group preventive maintenance time and maintenance cost by 150% compared with the single preventive maintenance; and propose the group preventive maintenance approach can be used for the maintenance of production equipment in semiconductor foundries in the future.