In order to alleviate the unreasonable sectorization in the terminal area, this paper studies the three-dimensional sectorization problem considering airspace function. Firstly, the objective of this paper is to reduce the workload of all sectors and to reduce the standard deviation of workload of different sectors on the premise of meeting the constraints of traditional sectorization. Then, combining the hierarchical clustering of vertical height and the horizontal segmentation of Voronoi diagram, the terminal area is divided into several three-dimensional cells and each cell is assigned a specific function by different types of flights. According to the optimization objective and considering the similarity of function, NSGA-II algorithm is used to combine all cells to group sectors. Finally, Shanghai terminal area is taken as an example to verify. The results show that compared with the current sectorization, the optimized sectorization can effectively reduce the workload of all sectors, and at the same time equalize the workload of different sectors.