A Deep Learning Approach for Flight Delay Prediction Through Time-Evolving Graphs

Cai, Kai-Yuan; Li, Yue; Fang, Yi-Ping; Zhu, Yongtian

doi:10.1109/tits.2021.3103502

Cited by 39 publications

(12 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The possibly first application of DL to delay prediction was proposed by Kim and co-authors, in which the sequences of departure and arrival flight delays of an airport were predicted using a Long Short-Term Memory network architecture, using input features like the delay of previous flights and the weather condition [24]. Numerous new studies have followed this initial work, mainly focusing on increasing the spectrum of information fed in the models: from micro-scale meteorological conditions [25]- [28], rea- VOLUME 4, 2016 sons of previous delays [29], airline and flights connection structure [26], [28], [30], [31], airport crowdedness [26], [32], to aircraft trajectories [33] and airspace structure [32]. The interested reader can refer to [34] for a review on the use of data analysis in the study of air transport delay.…”

mentioning

confidence: 99%

Assessing Identifiability in Airport Delay Propagation Roles Through Deep Learning Classification

2022

View full text Add to dashboard Cite

Delays in air transport can be seen as the result of two independent contributions, respectively stemming from the local dynamics of each airport and from a global propagation process; yet, assessing the relative importance of these two aspects in the final behaviour of the system is a challenging task. We here propose the use of the score obtained in a classification task, performed over vectors representing the profiles of delays at each airport, as a way of assessing their identifiability. We show how Deep Learning models are able to recognise airports with high precision, thus suggesting that delays are defined more by the characteristics of each airport than by the global network effects. This identifiability is higher for large and highly connected airports, constant through years, but modulated by season and geographical location. We finally discuss some operational implications of this approach.

show abstract

mentioning

confidence: 99%

Assessing Identifiability in Airport Delay Propagation Roles Through Deep Learning Classification

2022

View full text Add to dashboard Cite

show abstract

“…Cai et al considered the Spatio-temporal interactions in airport networks. They used graph-based convolutional neural networks to model the time evolution and periodic graph structure information in MAS networks [15] .…”

Section: Multi-airport Systemsmentioning

confidence: 99%

Multi-airport System Route Network Optimization Based on Differentiated Positioning

Wang

Guo²

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Air routes in China’s existing multi-airport system are unreasonable and homogeneous, leading to a significant gap in utilization between different airports and a waste of airport resources. One differentiated positioning model is proposed to optimize the air route network to solve the above problems. Balanced development of airports and passenger costs are considered in the model. Finally, the model is validated using the 2019 route network dataset in the Beijing-Tianjin-Hebei multi-airport system (BTH MAS). The optimization results for the multi-airport system show an average increase of 15% in airports utilization, an average increase of 6% in the differentiated realization rate, an 8% reduction in the average cost of passenger travel fees, and a 17% reduction in route repetition with no reduction in the number of connections to destination airports.

show abstract

“…Ai et al [39] employed a convolutional LSTM network to predict the flight delay, which has stronger time and space feature acquisition capability than traditional LSTM. Cai et al [40] proposed a flight delay prediction method based on GCNN to study the flight delay prediction problem from the perspective of network. Bisandu et al [41] presented a DL prediction method of flight delay to combine the social ski driver algorithm and conditional auto-regression risk value through regression quartile, which can help to select the correct network architecture and has better accuracy and less error compared with existing methods.…”

Section: Related Workmentioning

confidence: 99%

A sparrow search algorithm-optimized convolutional neural network for imbalanced data classification using synthetic minority over-sampling technique

Deng,

He,

Zhou

et al. 2023

Phys. Scr.

View full text Add to dashboard Cite

Due to the large span of flight delay, numerous influencing factors and large number of flights, the collected flight delay data is characterized by large-scale and imbalance, which is considered as one of the largest challenges in balanced flight delay data classification by using conventional classification algorithms. To solve this problem, an adaptive multi-class classification approach of imbalanced flight delay data using synthetic minority over-sampling technique (SMOTE)-based convolutional neural network with sparrow search algorithm, namely SSA-LTCNN is proposed in this paper. Firstly, the SMOTE is used to reduce the imbalance between the classes of the flight delay data. Then, the qualitative analysis method is employed to determine the key parameters of the CNN, which are optimized by using SSA to establish an optimized CNN model. Finally, the optimized CNN model with more effective feature representation ability is employed to extract discriminative features from the balanced flight delay data automatically. The obtained features in the last fully connected layer of the optimized CNN model are used for flight delay classification. The MIT-BIH Arrhythmia Data is selected to verify the effectiveness of the proposed SSA-LTCNN method and the confusion matrix, classification accuracy, precision, recall and F1 score are also used to evaluate the performance of the proposed SSA-LTCNN method, which can get better performance than RF, DT, LR, CNN and QRSCNN. The actual flight delay data is processed to verify the effectiveness of the proposed classification approach, which can achieve higher accuracy than KNN, NB, UCNN, CNN, SSA-LCNN and SSA-TCNN in different time periods. The experimental results show that the SSA-LTCNN can fast and effectively realize the classification of MIT-BIH Arrhythmia Data and flight delay data.

show abstract

A Deep Learning Approach for Flight Delay Prediction Through Time-Evolving Graphs

Cited by 39 publications

References 37 publications

Assessing Identifiability in Airport Delay Propagation Roles Through Deep Learning Classification

Assessing Identifiability in Airport Delay Propagation Roles Through Deep Learning Classification

Multi-airport System Route Network Optimization Based on Differentiated Positioning

A sparrow search algorithm-optimized convolutional neural network for imbalanced data classification using synthetic minority over-sampling technique

Contact Info

Product

Resources

About