Departure flight delay prediction model based on deep fully connected neural network

doi:10.11772/j.issn.1001-9081.2022010002

Journal of Computer Applications ›› 2022, Vol. 42 ›› Issue (10): 3283-3291.DOI: 10.11772/j.issn.1001-9081.2022010002

• Frontier and comprehensive applications • Previous Articles Next Articles

Departure flight delay prediction model based on deep fully connected neural network

Haiwen XU¹, Jiacai SHI², Teng WANG²

^1.School of Science，Civil Aviation Flight University of China，Guanghan Sichuan 618307，China
^2.College of Civil Aviation Safety Engineering，Civil Aviation Flight University of China，Guanghan Sichuan 618307，China

Received:2022-01-06 Revised:2022-04-25 Accepted:2022-04-27 Online:2022-10-14 Published:2022-10-10
Contact: Haiwen XU
About author:XU Haiwen， born in 1978， Ph. D. ， professor. His research interests include optimization theory and algorithms， transportation planning and management.
SHI Jiacai, born in 1996， M. S. candidate. His research interests include safety science， transportation planning and management.
WANG Teng, born in 1998， M. S. candidate. His research interests include safety science， transportation planning and management.
Supported by:
Fundamental Research Funds for Central Universities(J2021-057)

基于深度全连接神经网络的离港航班延误预测模型

徐海文¹, 史家财², 汪腾²

^1.中国民用航空飞行学院理学院，四川广汉 618307
^2.中国民用航空飞行学院民航安全工程学院，四川广汉 618307

通讯作者: 徐海文
作者简介:第一联系人：徐海文（1978—），男，山东菏泽人，教授，博士，主要研究方向：优化理论与算法、交通运输规划与管理； xuhaiwen_dream@163.com
史家财（1996—），男，山东济南人，硕士研究生，主要研究方向：安全科学、交通运输规划与管理
汪腾（1998—），男，广东江门人，硕士研究生，主要研究方向：安全科学、交通运输规划与管理。
基金资助:
中央高校基本科研业务费专项资金资助项目(J2021-057)

Abstract

Abstract:

Aiming at the problem that it is difficult to improve the accuracy of departure flight delay prediction， a departure flight delay prediction model based on Deep Fully Connected Neural Network （DFCNN） was proposed. Firstly， on the basis of considering flight information， airport weather and flight delay history， the influence of flight network structure on prediction model was considered. Secondly， experiments were carried out from three dimensions of activation function， input data item and delay time threshold to optimize and verify the model ability to suppress gradient dispersion and improve the learning performance. Finally， through adjusting the vertical expansion method of the number of neural network layers and the Dropout parameters of the random loss layers， the generalization ability of the model was improved. The results of experiments indicate that the prediction accuracy of the proposed model can be improved by 1.26 percentage points and 1.28 percentage points respectively after using tanh and Exponential Linear Unit （ELU） functions in the proposed model than using Rectified Linear Unit （ReLU）. After considering the flight network structure， the prediction accuracy calculated by the proposed model using ELU function is improved by 3.12 percentage points than without considering the flight network structure. When the Dropout parameters are adjusted， the loss value of the model is continuously reduced with 60 min time threshold. With a 5-layer hidden layer network and a Dropout parameter of 0.3， the prediction accuracy of 92.39% can be achieved by the proposed model. Therefore， the proposed model can make more accurate judgments on domestic flight delays.

Key words: deep learning, flight delay prediction, flight network, data fusion, model parameter

摘要：

针对提升离港航班延误预测精确度困难的问题，提出一种基于深度全连接神经网络（DFCNN）的离港航班延误预测模型。首先，在考虑航班信息、机场气象与航班延误历史的基础上，考虑航班网络结构对预测模型的影响；然后，从激活函数、输入数据项及延误时间阈值三个维度进行实验，以对模型抑制梯度弥散与提升学习表现能力的能力进行了优化与验证；最后，通过调控神经网络层数的纵向拓展方式与随机丢失层的Dropout参数，提升模型的泛化能力。实验结果表明：所提模型使用tanh、指数线性函数（ELU），预测精确度比使用线性整流函数（ReLU）分别提升了1.26、1.28个百分点；考虑航班网络结构后，所提模型采用ELU函数计算时，预测精确度比未考虑航班网络结构时提升了3.12个百分点；在时间阈值为60 min时，通过调控Dropout参数，模型的损失值不断降低；在5层隐含层网络和Dropout参数为0.3时，所提模型可以取得92.39%的预测精确度。因此，所提模型能够对国内航班延误做出较为准确的判断。

关键词: 深度学习, 航班延误预测, 航班网络, 数据融合, 模型参数

CLC Number:

TP391.4

Haiwen XU, Jiacai SHI, Teng WANG. Departure flight delay prediction model based on deep fully connected neural network[J]. Journal of Computer Applications, 2022, 42(10): 3283-3291.

徐海文, 史家财, 汪腾. 基于深度全连接神经网络的离港航班延误预测模型[J]. 《计算机应用》唯一官方网站, 2022, 42(10): 3283-3291.

Figures/Tables 18

Fig. 1 Every layer structure of neural network

Fig. 2 Calculation process of the proposed model

Tab. 1 Experimental data information of Chengdu Shuangliu Airport departure flights

研究对象	数据集	样本数比例/%
离港航班相关数据470 403条	训练集	60
	验证集	20
	测试集	20

Tab. 2 Datasets after merging data items

数据集合A	数据项				维度
数据集合A	$F m$	$W l$	$D n$	$N$	维度
A1	有	有	无	无	470 403×32
A2	有	有	有	无	470 403×36
A3	有	有	有	有	470 403×37

Tab. 2 Datasets after merging data items

数据集合A	数据项				维度
数据集合A	$F m$	$W l$	$D n$	$N$	维度
A1	有	有	无	无	470 403×32
A2	有	有	有	无	470 403×36
A3	有	有	有	有	470 403×37

Fig. 3 Schematic diagrams of standard neural network randomly stopping part of neuron nodes

Fig. 4 Loss value and accuracy change for different activation functions

Tab. 3 Experimental prediction results of different activation parameters

激活函数	损失值	精确度
tanh	0.218 6	0.830 0
ReLU	0.278 4	0.817 4
ELU	0.216 2	0.830 2

Tab. 4 Experimental prediction results of different input data

激活函数	输入数据	损失值	精确度
tanh	A1	0.218 6	0.830 0
	A2	0.219 2	0.829 7
	A3	0.182 7	0.858 8
ELU	A1	0.216 2	0.830 2
	A2	0.218 2	0.830 1
	A3	0.184 3	0.861 3

Fig. 5 Change of loss value of different input data under tanh and ELU functions

Fig. 6 Change of accuracy of different input data under tanh and ELU functions

Fig. 7 Change of loss value and accuracy with different thresholds under ELU function

Tab. 5 Experimental prediction results of different thresholds

TH/min	损失值	精确度
15	0.184 3	0.861 3
30	0.183 5	0.861 8
60	0.181 6	0.862 1

Tab. 6 Experimental prediction results of DFCNN with different hidden layers

隐含层层数	损失值	精确度
5	0.181 6	0.862 1
10	0.187 8	0.858 5
15	0.188 3	0.858 3

Fig. 8 Change of loss value and accuracy of DFCNN with different hidden layers

Tab. 7 Experimental prediction results of different Dropout parameters under different layers

隐含层层数	Dropout层参数p	损失值	精确度
5	0.3	0.109 3	0.923 9
	0.5	0.181 6	0.862 1
	0.7	0.213 5	0.847 9
10	0.3	0.114 8	0.919 4
	0.5	0.187 8	0.858 5
	0.7	0.226 2	0.843 6
15	0.3	0.121 2	0.914 1
	0.5	0.188 3	0.858 3
	0.7	0.224 8	0.842 1

Fig. 9 Change of loss value of different Dropout parameters under different layers

Fig. 10 Change of loss value of different Dropout parameters under different layers

Tab. 8 Comparison of prediction accuracy of different flight delay prediction models

模型	输入数据集	$P A c c$ /%
C4.5决策树	航班数据（国内）	82.00
贝叶斯网络	航班数据（国内）	88.00
随机森林	航班数据、气象数据（国内）	60.00
RNN	航班数据、气象数据（国外）	87.42
5LSTM	航班数据、气象数据（国外）	88.63
5ANN	航班数据、气象数据（国外）	88.64
改进5层DFCNN	A3（国内）	92.39

Tab. 8 Comparison of prediction accuracy of different flight delay prediction models

模型	输入数据集	$P A c c$ /%
C4.5决策树	航班数据（国内）	82.00
贝叶斯网络	航班数据（国内）	88.00
随机森林	航班数据、气象数据（国内）	60.00
RNN	航班数据、气象数据（国外）	87.42
5LSTM	航班数据、气象数据（国外）	88.63
5ANN	航班数据、气象数据（国外）	88.64
改进5层DFCNN	A3（国内）	92.39

References 33

1	朱江，黄建伟，亓洋洋，等. 基于超越对数成本函数的航空公司延误成本测算［J］. 物流科技， 2020， 43（12）：35-38， 45. 10.3969/j.issn.1002-3100.2020.12.010
	ZHU J， HUANG J W， QI Y Y， et al. Airline delay cost estimation based on Trans-log cost function［J］. Logistics Sci-Tech， 2020， 43（12）： 35-38， 45. 10.3969/j.issn.1002-3100.2020.12.010
2	Federal Aviation Administration. Cost of delay estimates 2019［EB/OL］. （2020-07-08）［2021-12-25］.. 10.2172/1768316
3	朱志国，周雨禾. 航班延误突发性群体事件疏导：基于改进的情绪感染模型［J］. 系统工程， 2018， 36（5）：79-84.
	ZHU Z G， ZHOU Y H. Evacuation of unexpected mass incidents caused by flight delays： based on an improved emotional contagion model［J］. Systems Engineering， 2018， 36（5）： 79-84.
4	谌婧娇. 基于Spark的决策树算法对航班延误预测研究［J］. 电脑知识与技术， 2021， 17（4）：217-219.
	CHEN J J. Prediction of flight delay based on decision tree algorithm of spark［J］. Computer Knowledge and Technology， 2021， 17（4）： 217-219.
5	高旗，初建宇，李印凤. 基于PSO-BP神经网络的终端区拥堵等级预测模型［J］. 航空计算技术， 2019， 49（6）：57-61. 10.3969/j.issn.1671-654X.2019.06.013
	GAO Q， CHU J Y， LI Y F. Prediction model of congestion level in terminal area based on PSO-BP neural network［J］. Aeronautical Computing Technique， 2019， 49（6）： 57-61. 10.3969/j.issn.1671-654X.2019.06.013
6	唐红，王栋，宋博，等. 基于非线性赋权XGBoost算法的航班延误分类预测［J］. 系统仿真学报， 2021， 33（9）：2261-2269. 10.16182/j.issn1004731x.joss.20-0372
	TANG H， WANG D， SONG B， et al. Classification of flight delay based on nonlinear weighted XGBoost［J］. Journal of System Simulation， 2021， 33（9）： 2261-2269. 10.16182/j.issn1004731x.joss.20-0372
7	宋捷，杨磊，胡明华，等. 基于深度学习的航班起降延误预测方法［J］. 航空计算技术， 2020， 50（3）：30-34. 10.3969/j.issn.1671-654X.2020.03.007
	SONG J， YANG L， HU M H， et al. Departure and landing delay prediction based on deep learning technique［J］. Aeronautical Computing Technique， 2020， 50（3）： 30-34. 10.3969/j.issn.1671-654X.2020.03.007
8	CHOI S， KIM Y J， BRICENO S， et al. Prediction of weather-induced airline delays based on machine learning algorithms［C］// Proceedings of the IEEE/AIAA 35th Digital Avionics Systems Conference. Piscataway： IEEE， 2016： 1-6. 10.1109/dasc.2016.7777956
9	李晓霞，吴薇薇，韩东，等. 基于聚类与贝叶斯网络的航班离港延误预测模型［J］. 哈尔滨商业大学学报（自然科学版）， 2020， 36（1）：110-113， 120.
	LI X X， WU W W， HAN D， et al. Flight departure delay prediction model based on clustering and Bayesian network［J］. Journal of Harbin University of Commerce （Natural Sciences Edition）， 2020， 36（1）： 110-113， 120.
10	王春政，胡明华，杨磊，等. 基于Agent模型的机场网络延误预测［J］. 航空学报， 2021， 42（7）：452-465. 10.7527/S1000-6893.2020.24604
	WANG C Z， HU M H， YANG L， et al. Airport network delay prediction based on Agent model［J］. Acta Aeronautica et Astronautica Sinica， 2021， 42（7）： 452-465. 10.7527/S1000-6893.2020.24604
11	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. 10.1016/j.procs.2016.09.321
12	吴仁彪，李佳怡，屈景怡. 基于双通道卷积神经网络的航班延误预测模型［J］. 计算机应用， 2018， 38（7）：2100-2106， 2112. 10.11772/j.issn.1001-9081.2018010037
	WU R B， LI J Y， QU J Y. Flight delay prediction model based on dual-channel convolutional neural network［J］. Journal of Computer Applications， 2018， 38（7）： 2100-2106， 2112. 10.11772/j.issn.1001-9081.2018010037
13	张舜尧，戴福青. 基于航空公司的航班计划优化与延误预测［J］. 科学技术与工程， 2021， 21（9）：3855-3860. 10.3969/j.issn.1671-1815.2021.09.060
	ZHANG S Y， DAI F Q. Flight plan optimization and delay prediction based on airlines［J］. Science Technology and Engineering， 2021， 21（9）： 3855-3860. 10.3969/j.issn.1671-1815.2021.09.060
14	谢华，李雨吟，胡明华，等. 基于BP神经网络的机场离港延误等级预测［J］. 航空计算技术， 2019， 49（3）：71-74. 10.3969/j.issn.1671-654X.2019.03.017
	XIE H， LI Y Y， HU M H， et al. Prediction of airport departure delay level based on BP neural network［J］. Aeronautical Computing Technique， 2019， 49（3）： 71-74. 10.3969/j.issn.1671-654X.2019.03.017
15	胡越，罗东阳，花奎，等. 关于深度学习的综述与讨论［J］. 智能系统学报， 2019， 14（1）：1-19. 10.11992/tis.201808019
	HU Y， LUO D Y， HUA K， et al. Overview on deep learning［J］. CAAI Transactions on Intelligent Systems， 2019， 14（1）： 1-19. 10.11992/tis.201808019
16	SRIVASTAVA N， HINTON G， KRIZHEVSKY A， et al. Dropout： a simple way to prevent neural networks from overfitting［J］. Journal of Machine Learning Research， 2014， 15： 1929-1958.
17	罗雄，邵荃，张海蛟，等. 航班延误引发的旅客群体性事件仿真研究［J］. 航空计算技术， 2014， 44（6）：25-29， 34. 10.3969/j.issn.1671-654X.2014.06.007
	LUO X， SHAO Q， ZHANG H J， et al. Research on simulation of passenger mass event derived from flight delay［J］. Aeronautical Computing Technique， 2014， 44（6）： 25-29， 34. 10.3969/j.issn.1671-654X.2014.06.007
18	尹晓丽. 大型机场航站楼旅客拥挤感知及控制策略研究［D］. 天津：中国民航大学， 2019：40-43.
	YIN X L. Research on passenger crowding perception and control strategy in large airport terminals［D］. Tianjin： Civil Aviation University of China， 2019： 40-43.
19	KIM J B. Introduction to the Montreal Convention 1999［J］. The Korean Journal of Air & Space Law and Policy， 2003， 17： 9-28.
20	CHENG B. A new era in the law of international carriage by air： from Warsaw （1929） to Montreal （1999）［J］. International and Comparative Law Quarterly， 2004， 53（4）： 833-859. 10.1093/iclq/53.4.833
21	付振宇，徐海文，傅强. 航班延误预测研究概述［J］. 科技与创新， 2020（3）：1-4.
	FU Z Y， XU H W， FU Q. Overview of flight delay prediction［J］. Science and Technology & Innovation， 2020（3）： 1-4.
22	吴仁彪，李佳怡，屈景怡. 融合气象数据的并行化航班延误预测模型［J］. 信号处理， 2018， 34（5）：505-512.
	WU R B， LI J Y， QU J Y. Parallel flight delay prediction model based on fusion of meteorological data［J］. Journal of Signal Processing， 2018， 34（5）： 505-512.
23	DOZAT T. Incorporating Nesterov momentum into Adam［EB/OL］（2016-02-19）［2021-12-25］..
24	周鑫. 全连接神经网络在FPGA上的实现与优化［D］. 合肥：中国科学技术大学， 2018：28-30.
	ZHOU X. Implementation and optimization of fully connected neural network on FPGA［D］. Hefei： University of Science and Technology of China， 2018： 28-30.
25	LI Y， FAN C X， LI Y， et al. Improving deep neural network with multiple parametric exponential linear units［J］. Neurocomputing， 2018， 301： 11-24. 10.1016/j.neucom.2018.01.084
26	GLOROT X， BENGIO Y. Understanding the difficulty of training deep feedforward neural networks［C］// Proceedings of the 13th International Conference on Artificial Intelligence and Statistics. New York： JMLR.org， 2010： 249-256.
27	徐海文，付振宇，傅强. 基于时效信息和深度学习的离港航班延误预测［J］. 科学技术与工程， 2020， 20（34）：14126-14132. 10.3969/j.issn.1671-1815.2020.34.026
	XU H W， FU Z Y， FU Q. The departure light delay prediction based on timely information and deep learning［J］. Science Technology and Engineering， 2020， 20（34）： 14126-14132. 10.3969/j.issn.1671-1815.2020.34.026
28	程俊华，曾国辉，鲁敦科，等. 基于Dropout的改进卷积神经网络模型平均方法［J］. 计算机应用， 2019， 39（6）：1601-1606. 10.11772/j.issn.1001-9081.2018122501
	CHENG J H， ZENG G H， LU D K， et al. Improved convolution neural network model averaging method based on Dropout［J］. Journal of Computer Applications， 2019， 39（6）： 1601-1606. 10.11772/j.issn.1001-9081.2018122501
29	程华，李艳梅，罗谦，等. 基于C4.5决策树方法的到港航班延误预测问题研究［J］. 系统工程理论与实践， 2014， 34（S1）：239-247. 10.12011/1000-6788(2014)s1-239
	CHENG H， LI Y M， LUO Q， et al. Study on flight delay with C4.5 decision tree based prediction method［J］. Systems Engineering—Theory and Practice， 2014， 34（S1）： 239-247. 10.12011/1000-6788(2014)s1-239
30	曹卫东，林翔宇. 基于贝叶斯网络的航班过站时间分析与延误预测［J］. 计算机工程与设计， 2011， 32（5）：1770-1772， 1776.
	CAO W D， LIN X Y. Flight turnaround time analysis and delay prediction based on Bayesian network［J］. Computer Engineering and Design， 2011， 32（5）： 1770-1772， 1776.
31	刘彤丹，许梦婷，陈舒伟，等. 基于天气影响的离场航班延误分析及预测［C/OL］// 2019世界交通运输大会论文集. ［2021-12-25］..
	LIU T D， XU M T， CHEN S W， et al. Analysis and prediction of departure flight delay based on weather impact［C/OL］// Proceedings of the World Transport Convention 2019. ［2021-12-25］..
32	KIN Y J， CHOI S， BRICENO S， et al. A deep learning approach to flight delay prediction［C］// Proceedings of the 35th IEEE/AIAA Digital Avionics Systems Conference. Piscataway： IEEE， 2016： 1-6. 10.1109/dasc.2016.7778092
33	屈景怡，叶萌，曹磊. 基于混合编码和长短时记忆网络的机场延误预测方法［J］. 信号处理， 2019， 35（7）：1160-1169.
	QU J Y， YE M， CAO L. Airport delay prediction method based on hybrid coding and LSTM network［J］. Journal of Signal Processing， 2019， 35（7）： 1160-1169.

[1]	Yexin PAN, Zhe YANG. Optimization model for small object detection based on multi-level feature bidirectional fusion [J]. Journal of Computer Applications, 2024, 44(9): 2871-2877.
[2]	Shunyong LI, Shiyi LI, Rui XU, Xingwang ZHAO. Incomplete multi-view clustering algorithm based on self-attention fusion [J]. Journal of Computer Applications, 2024, 44(9): 2696-2703.
[3]	Jing QIN, Zhiguang QIN, Fali LI, Yueheng PENG. Diagnosis of major depressive disorder based on probabilistic sparse self-attention neural network [J]. Journal of Computer Applications, 2024, 44(9): 2970-2974.
[4]	Xiyuan WANG, Zhancheng ZHANG, Shaokang XU, Baocheng ZHANG, Xiaoqing LUO, Fuyuan HU. Unsupervised cross-domain transfer network for 3D/2D registration in surgical navigation [J]. Journal of Computer Applications, 2024, 44(9): 2911-2918.
[5]	Yunchuan HUANG, Yongquan JIANG, Juntao HUANG, Yan YANG. Molecular toxicity prediction based on meta graph isomorphism network [J]. Journal of Computer Applications, 2024, 44(9): 2964-2969.
[6]	Yi LIU, Guoli YANG, Qibin ZHENG, Xiang LI, Yangsen ZHOU, Depeng CHEN. Architecture design of data fusion pipeline for unmanned systems [J]. Journal of Computer Applications, 2024, 44(8): 2536-2543.
[7]	Yuhan LIU, Genlin JI, Hongping ZHANG. Video pedestrian anomaly detection method based on skeleton graph and mixed attention [J]. Journal of Computer Applications, 2024, 44(8): 2551-2557.
[8]	Yanjie GU, Yingjun ZHANG, Xiaoqian LIU, Wei ZHOU, Wei SUN. Traffic flow forecasting via spatial-temporal multi-graph fusion [J]. Journal of Computer Applications, 2024, 44(8): 2618-2625.
[9]	Qianhong SHI, Yan YANG, Yongquan JIANG, Xiaocao OUYANG, Wubo FAN, Qiang CHEN, Tao JIANG, Yuan LI. Multi-granularity abrupt change fitting network for air quality prediction [J]. Journal of Computer Applications, 2024, 44(8): 2643-2650.
[10]	Yiqun ZHAO, Zhiyu ZHANG, Xue DONG. Anisotropic travel time computation method based on dense residual connection physical information neural networks [J]. Journal of Computer Applications, 2024, 44(7): 2310-2318.
[11]	Song XU, Wenbo ZHANG, Yifan WANG. Lightweight video salient object detection network based on spatiotemporal information [J]. Journal of Computer Applications, 2024, 44(7): 2192-2199.
[12]	Xun SUN, Ruifeng FENG, Yanru CHEN. Monocular 3D object detection method integrating depth and instance segmentation [J]. Journal of Computer Applications, 2024, 44(7): 2208-2215.
[13]	Zheng WU, Zhiyou CHENG, Zhentian WANG, Chuanjian WANG, Sheng WANG, Hui XU. Deep learning-based classification of head movement amplitude during patient anaesthesia resuscitation [J]. Journal of Computer Applications, 2024, 44(7): 2258-2263.
[14]	Huanhuan LI, Tianqiang HUANG, Xuemei DING, Haifeng LUO, Liqing HUANG. Public traffic demand prediction based on multi-scale spatial-temporal graph convolutional network [J]. Journal of Computer Applications, 2024, 44(7): 2065-2072.
[15]	Zhi ZHANG, Xin LI, Naifu YE, Kaixi HU. DKP： defending against model stealing attacks based on dark knowledge protection [J]. Journal of Computer Applications, 2024, 44(7): 2080-2086.

Departure flight delay prediction model based on deep fully connected neural network

基于深度全连接神经网络的离港航班延误预测模型

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 18

References 33

Related Articles 15

Recommended Articles

Metrics