A Moment-based Distributionally Robust Optimization Model for Air Traffic Flow Management

The Air Traffic Flow Management (ATFM) is considered as an effective method for air safety and efficiency guarantee. However, as one of the vital part of flight efficiency, fuel consumption is easily affected by external factors induced by various uncertainties such as severe weather conditions, which impact decision making process. Therefore, a distributionally robust mixed integer programming model for ATFM problem (DR-ATFM) is introduced in this paper to handle the uncertainty of fuel consumption while minimizing the total cost of fuel consumption, flight cancellation and flight delays. By exploiting the moment information of the fuel consumption data, a moment ambiguity set is constructed which characterizes the actual distribution of fuel consumption influctation. Based on this, an equivalent reformulation of DR-ATFM is derived to transform the problem into a mathematically solvable one, which is further solved through a cutting plane-based decomposition algorithm proposed in this paper. Finally, the effectiveness and robustness of the method are verified based on computational results of small-sized instances.