基于航班延误预测的多天气模式下时刻协调参数剖面研究

中国民航大学学报 ›› 2025, Vol. 43 ›› Issue (6): 24-30.

基于航班延误预测的多天气模式下时刻协调参数剖面研究

1. 中国民航大学空中交通管理学院，天津 300300； 2. 中国民用航空华北地区空中交通管理局终端管制室，北京 100621

收稿日期:2024-03-26 修回日期:2024-06-11 出版日期:2025-12-20 发布日期:2026-01-10
作者简介:高伟（1971— ），男，天津人，副教授，硕士，研究方向为交通运输规划与管理.
基金资助:
国家重点研发计划项目（2020YFB1600101）

Research on coordinated slot parameter profiles based on flight delay

prediction under multiple weather patterns

1. College of Air Traffic Management, CAUC, Tianjin 300300, China; 2. Approach Control Unit, North China Regional

Air Traffic Management Bureau of CAAC, Beijing 100621, China

Received:2024-03-26 Revised:2024-06-11 Online:2025-12-20 Published:2026-01-10

摘要/Abstract

摘要：

构造符合正常性期望的繁忙机场 18~24 h 合理时刻协调参数剖面用于公布容量和时刻换季是尚未解决的

民航时刻管理难题。本文使用机场航班和天气历史数据，通过 K-means 聚类及偏最小二乘回归建立时刻

结构的回归预测模型，利用集成学习预测航班延误水平。结果表明：随机森林在回归和预测方面都呈现较

好的效果，并能结合航班延误预测得到时刻协调参数剖面的上限与下限作为时刻协调参数区间；将结果进

行仿真验证，使得在区间内的航班架次安排满足小于 15 min 的平均延误时间水平，最终给出建议的时刻

协调参数。本文可为相关部门的时刻管理，不同战略战术时期的空中交通流量管理，以及机场和航空公司

评估延误风险、调整时刻安排、配置运力和保障资源，提供精细化的辅助决策信息。

关键词:

空中交通, 多天气模式, 航班时刻, 航班延误, 集成学习

Abstract:

Constructing a reasonable 18—24 h coordinated slot parameter profile for busy airports that aligns with normality

expectations, used for publishing capacity and facilitating seasonal slot adjustments, remains an unresolved issue

in aviation slot management. This paper develops a regression prediction model for slot structure using historical

flight and weather data, employing K-means clustering and partial least squares regression, while applying

ensemble learning to forecast flight delay levels. The results show that random forests exhibit good performance in

both regression and prediction, and can combine flight delay predictions to obtain the upper and lower limits of

the coordinated slot parameter profile as the slot coordination parameter interval. The results are verified through

simulation, ensuring that the flight schedule within this interval maintains an average delay level of less than 15

min, leading to the proposal of recommended slot coordination parameters. This paper provides refined decision support information for relevant departments in slot management and air traffic flow management during different

strategic and tactical periods, as well as for airports and airlines in evaluating delay risks, adjusting slot arrangements, allocating capacity, and planning support resources.

Key words:

air traffic, multiple weather patterns, flight slots, flight delay, ensemble learning

中图分类号:

V352

高伟, 逄丁荧, .

基于航班延误预测的多天气模式下时刻协调参数剖面研究

[J]. 中国民航大学学报, 2025, 43(6): 24-30.

GAO Wei , PANG Dingying, .

Research on coordinated slot parameter profiles based on flight delay

prediction under multiple weather patterns

[J]. Journal of Civil Aviation University of China, 2025, 43(6): 24-30.

参考文献 (1)

[1]	中国民用航空局. 民航局关于印发《民航航班时刻管理办法》的通知[J]. 中华人民共和国国务院公报, 2018(19): 79-86.
[2]	KHAN A M. Airport slot controls[M]//Handbook of Transport Systems
	and Traffic Control. Bradford, West Yorkshire: Emerald Group Publishing Limited, 2001: 579-590.
[3]	王艳军, 水笑雨, 王梦尹. 机场航班时刻资源管理研究进展[J]. 北京航空航天大学学报, 2024, 50(4): 1065-1076.
[4]	呙常冠. 空中交通综合流量预测系统[J]. 民航管理, 2016(11): 52-
	57.
[5]	杨波. 从同一航班号的时刻差异问题谈航班时刻协调参数管
理[J]	中国民用航空, 2019(1): 25-28.
[6]	GILLEN D, JACQUILLAT A, ODONI A R. Airport demand management: the operations research and economics perspectives and potential
	synergies[J]. Transportation Research Part A: Policy and Practice, 2016,
	94: 495-513.
[7]	交通运输部. 航班正常管理规定[J]. 中华人民共和国国务院公报,
	2021(28): 22-31.
[8]	OFFERMAN H. Simulation to support the airport stakeholder decision making process[J]. Air & Space Europe, 2001, 3(1/2): 60-67.
[9]	周丰. 基于 K-means 聚类算法的航材故障预警方法研究[J]. 民航学报, 2022, 6(2): 90-92, 103.

	[10] 涂钰. 基于贝叶斯优化集成学习算法的航班延误预测融合模型研究[D]. 成都: 西南交通大学, 2021.


	[11] 王辉, 张文杰, 刘杰, 等. 基于分类回归决策树算法的航班延误预测模型[J]. 中国民航大学学报, 2022, 40(3): 35-40.


	[12] 刘博, 叶博嘉, 田勇, 等. 航班延误预测方法研究综述[J]. 航空计算技术, 2019, 49(6): 124-128.


	[13] STEFANOVICˇ P, 譒TRIMAITIS R, KURASOVA O. Prediction of flight time deviation for Lithuanian Airports using supervised machine learning model[J]. Computational Intelligence and Neuroscience, 2020, 2020


(1)	: 8878681.


	[14] RODR魱GUEZ-SANZ 魣, COMENDADOR F G, VALD魪S R A, et al. Assessment of airport arrival congestion and delay: prediction and reliability[J]. Transportation Research Part C: Emerging Technologies, 2019,


	98: 255-283.


	[15] KHANMOHAMMADI S, TUTUN S, KUCUK Y. A new multilevel input


	layer artificial neural network for predicting flight delays at JFK airport[J]. Procedia Computer Science, 2016, 95: 237-244.


	[16] 刘中祥, 王欣. 基于随机森林回归的航班延误预测[J]. 现代计算机, 2019, 25(15): 20-24.


	[17] 曹卫东, 张金迪, 刘晨宇. 基于 ST-LightGBM 的机场离港航班延误预测[J]. 陕西科技大学学报, 2023, 41(4): 166-172.


	[18] 徐海文, 汪腾. 考虑航路网络结构的离场航班延误预测模型[J].科学技术与工程, 2023, 23(11): 4734-4744.