For the aircraft cluster scheduling problem in a limited space, the paper models the overall operation process first, then converts the model to a typical scheduling model, for which heuristic rules and an optimization algorithm are presented. The algorithm is divided into upper and lower levels. The upper layer is responsible for selecting parking space and take-off position in the limited area. The lower layer is responsible for the scheduling work of each single aircraft and returns the scheduling result to the upper layer, which adaptively adjusts the result accordingly. Then, the lower layer algorithm is recalled to verify whether the adjustments of stand and takeoff position can improve the quality of the solution. The result of the scheduling scheme is demonstrated. Compared withmanual arrangement, the method is more advantageous in both solving-time and solution quality. This method is of great significance to improve the efficiency of aircraft sorties.