Scientifically establishing diversified peak-valley configuration schemes of flight slot coordination parameter rep-resents a potent strategy for enhancing the flexibility of flight slot scheduling and alleviating planned primary delays at busy airport. This paper proposes a general method for the generic peak-valley configuration of flight slot coordi-nation parameter. Firstly,the structure of airport slot fire-break is defined,and the determination process of a generic peak-valley configuration scheme of flight slot coordination parameter is introduced. Secondly,a schedule-based prediction model for airport primary delay is developed,along with the design of rules generating peak-valley feature parameters of flight slot and expected delay datasets. Finally,the synthetic minority oversampling technique (SMOTE) and edited nearest neighbors (ENN) algorithm are then employed to address the issue of imbalanced dataset category,and various decision tree classification algorithms are applied to summarize the generic peak-valley con-figuration pattern set of airport tree-structured slot coordination parameters under a certain level of delay. Taking Shanghai Pudong International Airport as an example,the prediction accuracy of the primary delay model for airport arrival and departure,as well as the accuracy of extracting the airport peak-valley configuration combination of flight slot coordination parameter are verified,and the pattern set of tree-structured peak-valley configuration com-bination under certain delay level is established. This paper can offer novel insights and methods for the refined set-ting of airport slot coordination parameter,and providing theoretical support for the optimization and configuration of flight slot.
flight slot coordination parameterslot fire-breakdata-drivendecision tree algorithmflight management
维普
万方数据
