Application of a predictivefuel consumption modelusing machine learning andrandom forest for a Colombiancommercial airline
DOI:
https://doi.org/10.18667/cienciaypoderaereo.826Keywords:
Artificial intelligence, evaluation metrics, feature engineering, fuel forecasting, machine learning, random forestAbstract
In the aeronautical industry, JET A-1 hydrocarbon production costs experience an increase of approximately 5% weekly over the years, which has resulted in a rise in expenses for air operators. This research proposes to create a fuel consumption prediction model for regional flights, based on data collected from flights of a Colombian airline during the period 2018-2019. To create the predictive model, the Sci-Kit Learn library of the Python programming language was used, and the problem was approached from an ‘inverse problem’ perspective. Feature engineering was then carried out to improve the quality of the obtained data set and allow for greater prediction accuracy. The prediction model was implemented for the variables identified as dependent and independent, and finally, its performance was evaluated using the metrics of Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE). The results indicate that the model is capable of predicting fuel consumption, with low errors in quantitative values.
Downloads
References
Baevski, A., Hsu, W.-N., Xu, Q., Babu, A., Gu, J. y Auli, M. (2022). data2vec: A General Framework for Self-Supervised Learning in Speech, Vision and Language. Proceedings of Machine Learning Research, 162, 1298-1312. https://proceedings.mlr.press/v162/baevski22a.html
Cándaño-Viñas, L. y Gómez, M. (2023). El impacto de la covid-19 en la demanda turística internacional. ININEE CIENCIA: Revista de Investigación Científica, 1(1), 23-38. https://doi.org/10.33110/inineeciencia.v1i1.5
Cheung, J. C. H., Wells, C. A. y Steele, E. C. C. (2023). Evaluation of Methods of Estimating Time-Optimal Flight Routes in a Changing Climate. Meteorological Applications, 30(2). https://doi.org/10.1002/met.2121
Duarte, J. M. y Berton, L. (2023). A Review of Semi-Supervised Learning for Text Classification. Artificial Intelligence Review, 56(9), 9401-9469. https://doi.org/10.1007/s10462-023-10393-8
Espinosa-Zúñiga, J. J. (2020). Aplicación de algoritmos Random Forest y xgBoost en una base de solicitudes de tarjetas de crédito. Ingeniería, Investigación y Tecnología, 21(3), 1-16. https://doi.org/10.22201/fi.25940732e.2020.21.3.022
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M. y Duchesnay, É. (2011). Scikitlearn: Machine Learning in Python. Journal of Machine Learning Research, 12(85), 2825-2830. https://tinyurl.com/5r9a3xsy
Fieri, B. y Suhartono, D. (2023). Offensive Language Detection Using Soft Voting Ensemble Model. mendel: Soft Computing Journal, 29(1), 1-6. https://doi.org/10.13164/mendel.2023.1.001
Gallet, A., Rigby, S., Tallman, T. N., Kong, X., Hajirasouliha, I., Liew, A., Liu, D., Chen, L., Hauptmann, A. y Smyl, D. (2022). Structural Engineering from an Inverse Problems Perspective. Proceedings of the Royal Society, 478(2257). http://doi.org/10.1098/rspa.2021.0526
Girón-Girón, L. Á (2023). Estimación de demanda de transporte aéreo de pasajeros desde un enfoque de redes para aeropuertos tipo hub: caso Aeropuerto El Dorado Bogotá (tesis de maestría, Universidad Nacional de Colombia). Repositorio institucional UNAL. https://tinyurl.com/56beb5c6
Hassan, T. H., Sobaih, A. E. E. y Salem, A. E. (2021). Factors Affecting the Rate of Fuel Consumption in Aircrafts. Sustainability, 13(14). https://doi.org/10.3390/su13148066
Hodson, T. O. (2022). Error cuadrático medio (rmse) o error absoluto medio (mae): cuándo usarlos o no. Geoscientific Model Development, 15(14), 5481-5487. https://doi.org/10.5194/gmd-15-5481-2022
International Air Transport Association (iata). (2024, abril 5). Jet Fuel Price Monitor [en línea]. https://tinyurl.com/n27pdcaz
Janiesch, C., Zschech, P. y Heinrich, K (2021). Machine Learning and Deep Learning. Electron Markets, 31, 685-695. https://doi.org/10.1007/s12525-021-00475-2
Lozano, C., Melo, P., Bermúdez, J., González, S. y Fuentes, S. (2023). Integración de algoritmos de inteligencia artificial en función del pronóstico de consumo de combustible en aeronaves ERJ 145 de una aerolínea colombiana [ponencia]. VII Congreso Argentino de Ingeniería Aeronáutica (CAIA 7), La Plata, Argentina, 29 de noviembre al 1 de diciembre de 2023. https://tinyurl.com/2sr8kmp6
Márquez Díaz, J. (2020). Inteligencia artificial y Big Data como soluciones frente a la covid-19. Revista de Bioética y Derecho, (50), 315-331. https://tinyurl.com/mr3mvac5
Morrison, J. K. D., Bonnefoy, P. A. y Hansman, R. J. (2010). Investigation of the Impacts of Effective Fuel Cost Increase on the US Air Transportation Network and Fleet [ponencia]. 10th AIAA Aviation Technology, Integration, and Operations Conference, 13-15 de septiembre de 2010. https://tinyurl.com/vyjhrd9y
Mouillet, V., Nuic, A., Casado, E. y López-Leonés, J. L. (2018). Evaluación de la aplicabilidad de un modelo moderno de rendimiento de aeronaves para la optimización de trayectorias. IEEE/AIAA 37th Digital Avionics Systems Conference (DASC), 23-27 de septiembre de 2018. https://doi.org/10.1109/dasc.2018.8569732
Orellano-Lasprilla, J. L. (2023). Análisis fluido dinámico computacional de un mezclador estático para combustible de aviación JET A-1 y biodiesel (tesis de maestría, Universidad ECCI). Repositorio institucional ECCI. https://doi.org/10.1109/DASC.2018.8569732
Panda, M., Mousa, A. A., & Hassanien, A. E. (2021). Developing an efficient feature engineering and machine learning model for detecting IOT-botnet cyber attacks. IEEE Access, 9, 91038–91052. https://doi.org/10.1109/access.2021.3092054
Pizlo, Z. (2001). Perception Viewed as an Inverse Problem. Vision Research, 41(24), 3145-3161. https://doi.org/10.1016/S0042-6989(01)00173-0
Rativa-Sáenz, E. A. (2022). Análisis de series de tiempo para la estimación del consumo combustible en aeronaves ATR42 para aerolínea colombiana (tesis de especialización, Fundación Universitaria Los Libertadores). https://tinyurl.com/3eaay68a
Rosenow, J., Fricke, H., Luchkova, T. y Schultz, M. (2019). Impact of Optimised Trajectories on Air Traffic Flow Management. The Aeronautical Journal, 123(1260), 157-173. https://doi.org/10.1017/aer.2018.155
Sardi, A., Sorano, E., Cantino, V. y Garengo, P. (2020). Big Data and Performance Measurement Research: Trends, Evolution and Future Opportunities. Measuring Business Excellence, 27(4), 531-548. https://doi.org/10.1108/mbe-06-2019-0053
Schröer, C., Kruse, F. y Gómez, J. M. (2021). A Systematic Literature Review on Applying crisp-dm Process Model. Procedia Computer Science, 181, 526-534. https://doi.org/10.1016/J.PROCS.2021.01.199
Singh, V., & Sharma, S. K. (2015). Fuel consumption optimization in air transport: A review, classification, critique, simple meta-analysis, and future research implications. European Transport Research Review, 7(2). https://doi.org/10.1007/s12544-015-0160-x
Ying, X. (2019). An Overview of Overfitting and its Solutions. Journal of Physics: Conference Series, 1168. https://doi.org/10.1088/1742-6596/1168/2/022022
Yu, L., Zhao, Y., Tang, L. y Yang, Z. (2019). Online Big Data-Driven Oil Consumption Forecasting with Google Trends. International Journal of Forecasting, 35(1), 213-223. https://doi.org/10.1016/j.ijforecast.2017.11.005
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Escuela de Postgrados de la Fuerza Aérea Colombiana
This work is licensed under a Creative Commons Attribution 4.0 International License.
Assignment of Copyrights
Authors assign Ciencia y Poder Aéreo journal the exclusive rights (reproduction, distribution, public communication, and transformation) to exploit and commercialize their work, in whole or in part, in all the formats and modalities of present or future exploitation, in all languages, throughout the life of the work and throughout the world.
All contents published in Ciencia y Poder Aéreo journal are licensed under a Creative Commons Attribution 4.0 International License, whose complete information is available at http://creativecommons.org/licenses/by/4.0/
Under the terms of this license, users are free to download, print, extract, archive, distribute and publicly communicate the content of articles, provided that proper credit is granted to authors and Ciencia y Poder Aéreo, scientific journal of the Graduate School of the Colombian Air Force. Except when otherwise indicated, this site and its contents are licensed under a Creative Commons Attribution 4.0 International License.
For other uses not considered under this license it is required to contact the Director or the Editor of the journal at the e-mail address cienciaypoderaereo1@gmail.com.
The Graduate School of the Colombian Air Force and this publication are not responsible for the concepts expressed in the articles, including the metadata or the affiliation stated by authors. This is the full responsibility of the authors.
