Flight Delay Prediction with Airport Traffic Density Data from an Aviation Risk Management Perspective

Abstract

Flight delays are significantly important in risk management for the aviation industry, impacting airline operations, passenger satisfaction, and air traffic management. While existing studies primarily focus on weather-related factors in flight delay prediction, this study explores the influence of airport traffic density on delays from an aviation risk management perspective. Using data mining techniques, the study integrates airport traffic and en-route delay datasets from EUROCONTROL to develop predictive models for delay estimation. The methodology follows a structured approach, including data preprocessing, feature engineering, clustering, and predictive modeling using the Random Forest algorithm. The findings indicate that airport traffic density is a critical predictor of delays, alongside seasonal and regional factors. Regression analysis highlights a strong correlation between congestion levels and delay severity, particularly in peak travel periods. The clustering results reveal four distinct delay patterns, reflecting variations in operational disruptions due to equipment failures and adverse weather conditions. The Random Forest model demonstrates high predictive accuracy, with low error rates confirming its robustness for delay estimation. This study contributes to aviation risk management by providing data-driven insights into flight delays and offering strategic decision-making tools for airline and airport operators. The results emphasize the need for proactive delay mitigation strategies, such as improved airspace allocation and enhanced maintenance processes. Future research could extend this approach by incorporating additional delay factors, such as incident-related disruptions, to further enhance predictive capabilities. By integrating operational data and advanced analytics, this study presents a novel framework for improving delay forecasting and optimizing flight operations.

Keywords

• Aviation
• Flight delays
• Predictive modeling
• Random forest algorithm
• Risk management

Havacılık Risk Yönetimi Perspektifinden Havaalanı Trafik Yoğunluğu Verileriyle Uçuş Gecikmesi Tahmini

Abstract

Uçuş gecikmeleri havacılık sektöründe risk yönetiminde önemli bir öneme sahiptir ve havayolu operasyonlarını, yolcu memnuniyetini ve hava trafiği yönetimini etkiler. Mevcut çalışmalar öncelikli olarak uçuş gecikmesi tahmininde hava durumuyla ilgili faktörlere odaklanırken, bu çalışma havaalanı trafik yoğunluğunun gecikmeler üzerindeki etkisini havacılık risk yönetimi perspektifinden araştırmaktadır. Veri madenciliği tekniklerini kullanarak çalışma, gecikme tahmini için tahmini modeller geliştirmek üzere EUROCONTROL'den havaalanı trafiği ve rota gecikmesi veri kümelerini entegre etmektedir. Metodoloji, veri ön işleme, özellik mühendisliği, kümeleme ve Random Forest algoritmasını kullanarak tahmini modellemeyi içeren yapılandırılmış bir yaklaşımı takip etmektedir. Bulgular, havaalanı trafik yoğunluğunun mevsimsel ve bölgesel faktörlerin yanı sıra gecikmelerin kritik bir tahmincisi olduğunu göstermektedir. Regresyon analizi, özellikle yoğun seyahat dönemlerinde tıkanıklık seviyeleri ve gecikme şiddeti arasında güçlü bir korelasyon olduğunu vurgulamaktadır. Kümeleme sonuçları, ekipman arızaları ve olumsuz hava koşulları nedeniyle operasyonel kesintilerdeki değişiklikleri yansıtan dört farklı gecikme modelini ortaya koymaktadır. Random Forest modeli, gecikme tahmini için sağlamlığını doğrulayan düşük hata oranlarıyla yüksek tahmini doğruluk göstermektedir. Bu çalışma, uçuş gecikmelerine ilişkin veri odaklı tahminler sağlayarak ve havayolu ve havaalanı operatörleri için stratejik karar alma araçları sunarak havacılık risk yönetimine katkıda bulunmaktadır. Sonuçlar, iyileştirilmiş hava sahası tahsisi ve geliştirilmiş bakım süreçleri gibi proaktif gecikme azaltma stratejilerine olan ihtiyacı vurgulamaktadır. Gelecekteki araştırmalar, tahmin yeteneklerini daha da geliştirmek için olayla ilgili kesintiler gibi ek gecikme faktörlerini dahil ederek bu yaklaşımı genişletebilir. Operasyonel verileri ve gelişmiş analitiği entegre ederek, bu çalışma gecikme tahminini iyileştirmek ve uçuş operasyonlarını optimize etmek için yeni bir çerçeve sunmaktadır.

Keywords

• Havacılık
• Uçuş gecikmeleri
• Tahmine dayalı modelleme
• Rastgele orman algoritması
• Risk yönetimi

References

1. Ans Performance. (2024). Airport traffic dataset. EUROCONTROL. Retrieved October 25, 2024, from https://ansperformance.eu/reference/dataset/airport-traffic/
2. Ans Performance. (2024). En-route ATFM delay dataset. EUROCONTROL. Retrieved October 25, 2024, from https://ansperformance.eu/reference/dataset/en-route-atfm-delay-fir/
3. Binias, B., Myszor, D., Palus, H., & Cyran, K. A. (2020). Prediction of pilot's reaction time based on EEG signals. Frontiers in Neuroinformatics, 14.
4. Cai, K. Q., Li, Y., Fang, Y. P., & Zhu, Y. B. (2022). A deep learning approach for flight delay prediction through time-evolving graphs. IEEE Transactions on Intelligent Transportation Systems, 23(8), 11397–11407.
5. Dursun, Ö. O. (2023). Air-traffic flow prediction with deep learning: a case study for Diyarbakır airport. Journal of Aviation, 7(2), 196-203.
6. Esmaeilzadeh, E., & Mokhtarimousavi, S. (2020). Machine learning approach for flight departure delay prediction and analysis. Transportation Research Record, 2674(8), 145–159.
7. Fernandes, N., Moro, S., Costa, C. J., & Aparício, M. (2020). Factors influencing charter flight departure delay. Research in Transportation Business and Management, 34.
8. Gui, G., Liu, F., Sun, J., Yang, J., Zhou, Z., & Zhao, D. (2019). Flight delay prediction based on aviation big data and machine learning. IEEE Transactions on Vehicular Technology, 69(1), 140–150.

9. Jiang, Y. S., Niu, S. T., Zhang, K., Chen, B. W., Xu, C. T., Liu, D. H., & Song, H. B. (2022). Spatial-temporal graph data mining for IoT-enabled air mobility prediction. IEEE Internet of Things Journal, 9(12), 9232–9240.
10. Jiang, Y., Liu, Y., Liu, D., & Song, H. (2020). Applying machine learning to aviation big data for flight delay prediction. In 2020 IEEE International Conference on Dependable, Autonomic and Secure Computing,
11. International Conference on Pervasive Intelligence and Computing (pp. 665–672) IEEE.
12. Liu, M., Guo, C., & Xu, L. (2024). An interpretable automated feature engineering framework for improving logistic regression. Applied Soft Computing, 153, 111269.
13. Luo, Q., Zhang, L., Xing, Z., Xia, H., & Chen, Z. X. (2021). Causal discovery of flight service process based on event sequence. Journal of Advanced Transportation, 2021(1), 2869521.
14. Ma, X. Y., He, Z., Yang, P. F., Liao, X. Y., & Liu, W. (2024). Agent-based modelling and simulation for life-cycle airport flight planning and scheduling. Journal of Simulation, 18(1), 15–28.
15. Qu, J., Zhao, T., Ye, M., Li, J., & Liu, C. (2020). Flight delay prediction using deep convolutional neural network based on fusion of meteorological data. Neural Processing Letters, 52(2), 1461–1484.
16. Reitmann, S., & Schultz, M. (2022). An adaptive framework for optimization and prediction of air traffic management (sub-)systems with machine learning. Aerospace, 9(2).
17. Sarveswararao, V., Ravi, V., & Vivek, Y. (2023). ATM cash demand forecasting in an Indian bank with chaos and hybrid deep learning networks. Expert Systems with Applications, 211, 118645.
18. Schultz, M., Reitmann, S., & Alam, S. (2021). Predictive classification and understanding of weather impact on airport performance through machine learning. Transportation Research Part C: Emerging Technologies, 131.
19. Truong, D. (2021). Using causal machine learning for predicting the risk of flight delays in air transportation. Journal of Air Transport Management, 91.
20. Wang, C., Hu, M., Yang, L., & Zhao, Z. (2021). Prediction of air traffic delays: An agent-based model introducing refined parameter estimation methods. PLoS One, 16(4), 0249754.
21. Zeng, W. L., Ren, Y. M., Wei, W. B., & Yang, Z. (2021). A data-driven flight schedule optimization model considering the uncertainty of operational displacement. Computers & Operations Research, 133.
22. Zhang, H., Song, C. Y., Zhang, J., Wang, H., & Guo, J. L. (2021). A multi-step airport delay prediction model based on spatial-temporal correlation and auxiliary features. IET Intelligent Transport Systems, 15(7), 916–928.
23. Zhang, K., Jiang, Y., Liu, D., & Song, H. (2020). Spatio-temporal data mining for aviation delay prediction. In 2020 IEEE 39th International Performance Computing and Communications Conference (IPCCC) (pp. 1–7). IEEE.
24. Zhang, Y. N., Lu, Z., Wang, J. X., & Chen, L. (2023). FCM-GCN-based upstream and downstream dependence model for air traffic flow networks. Knowledge-Based Systems, 260.
25. Zhao, Z., Yuan, J., & Chen, L. (2024). Air traffic flow management delay prediction based on feature extraction and an optimization algorithm. Aerospace, 11(2), 168.
26. Zhu, D., Wang, H. W., & Tan, X. H. (2024). Mining delay propagation causality within an airport network from historical data. Aerospace, 11(7).
27. Zhu, Q., Chen, S., Guo, T., Lv, Y., & Du, W. (2024). A spatio-temporal approach with self-corrective causal inference for flight delay prediction. IEEE Transactions on Intelligent Transportation Systems.