The monthly maintenance scheduling is an important work for power dispatching institution. Its scientificity and reasonability directly affect the security of the power system. This paper defines the conception of the risk overlap and proposes the quantification method of the degree of risk overlap. The database index technology is used to find the set of objects affected by failure equipment. The transfer of the risk among objects with different voltage grades is also taken into consideration. On the basis of the failure index and dependency index of the sources of the object, the risk overlap can be judged when two or several works are carried out at the same time. The minimization of risk overlap is included in the objective function. The minimization of loss of power and equilibrium of workload are regarded as constraint conditions. The artificial fish-swarm algorithm is applied to optimize the maintenance scheduling. Experimental results show that the method in this paper can minimize the degree of risk overlap of the maintenance scheduling and realize the collaborative optimization with multi-voltage grade. This work is supported by National Natural Science Foundation of China (No. 51707111).