A data-driven flight schedule optimization model considering the uncertainty of operational displacement
A data-driven flight schedule optimization model considering the uncertainty of operational displacement
复制标题
DOI：
10.1016/j.cor.2021.105328
复制
发表时间：
2021-05-05
影响因子：
4.6
通讯作者：
Yang, Zhao
中科院分区：
文献类型：
--
作者：
Zeng, Weili;Ren, Yumeng;Yang, Zhao
时段分配机制旨在从策略角度配合航班需求和机场资源。然而，目前的研究主要集中在航空公司的利益上，忽略了导致航班原发延误的影响因素，使得航班时刻的时空分布不合理。本文提出了一种数据驱动的方法，通过考虑运营效率和航空公司的利益，在战略层面上减少运营延误。首先从历史运行数据中挖掘出实际执行时间与计划执行时间之间的位移概率分布。然后，我们开发了一个模型，最终目标是提高准点率和减少实际的操作延迟，最大限度地减少总的操作位移。该模型除了考虑机场的基本运行限制外，还引入了机场周边终端空域的走廊容量，在一定程度上减少了走廊流量控制带来的延误。该模型被应用于中国杭州萧山国际机场。实验结果表明，优化后的航班时刻表能显著减少航班延误，符合机场运行限制，保持航班衔接。
The slot allocation mechanism aims to match flight demand and airport resources from a strategic perspective. However, current research mainly focused on airlines' interests, ignoring the influencing factors that lead to primary delays, which makes the temporal and spatial distribution of flight schedules unreasonable. This paper proposes a data-driven approach to reduce operational delays at a strategic level by considering operational efficiency and airline interests. The displacement probability distribution between the actual execution time and the scheduled time is first mined from the historical operation data. We then develop a model with the ultimate objective of improving the punctuality rate and reducing the actual operational delays by minimizing the total operational displacement. In addition to considering the basic operational restrictions of airports, the model also introduces the corridor capacity of the terminal airspace surrounding an airport, reducing the delay caused by the corridor flow control to a certain extent. The proposed model is applied to Hangzhou Xiaoshan International Airport in China. The experimental results suggest that the optimized flight schedule can significantly reduce flight delay, conforms to airport operational restrictions, and maintains flight connectivity.