Issue |
ITM Web Conf.
Volume 48, 2022
The 4th International Conference on Computing and Wireless Communication Systems (ICCWCS 2022)
|
|
---|---|---|
Article Number | 03005 | |
Number of page(s) | 9 | |
Section | Computer Science, Intelligent Systems and Information Technologies | |
DOI | https://doi.org/10.1051/itmconf/20224803005 | |
Published online | 02 September 2022 |
Impact of Aircraft Performance and Time of the Day on Flight Arrival Delays Prediction in the United States: a Machine Learning Classification
1 Laboratory of Mathematics, Computer and Engineering sciences. Mathematics and Computer Science Department
2 Faculty of Science and Techniques Hassan First University of Settat. 26000 Settat, Morocco
* e-mail: h.alla@uhp.ac.ma
** e-mail: lahcen.moumoun@uhp.ac.ma
*** e-mail: youssef.balouki@uhp.ac.ma
The excessive growth of air traffic, with the limited airspace and airports capacity, results in a flight demand-capacity imbalance leading to air traffic delays. This paper explores the factors associated with delay in both microscopic and macroscopic ways. The aim is to develop a model which analyzes and predicts the occurrence of flight arrival delays using US domestic flight data for the year 2018. It will provide passengers, airlines and airport managers with reliable flight arrival schedules, and consequently reduce economic losses and enhance passengers trust. Beside database features, the proposed model is to the best of our knowledge the first attempt to predict flight arrival delays using three new features which are contributive factors to delays: Departure Time and Arrival Time of the day in which the flight was performed (Early morning, late morning, noon, afternoon, evening or night) and model of aircraft. Four Machine Learning classifiers namely Random Forest, Decision Trees, K-Nearest Neighbors and Naive Bayes were used. In order to find the best parameters of each algorithm, we implemented Grid Search technique. The performance of each classifier was compared in terms of hyperparameters tuning, classification metrics and features description. The experimental results showed that the proposed system was able to predict flight arrival delays with the best Random Forest accuracy of 0.9356 and a higher number of correctly classified flights. To prove the importance of our findings, we compared our model to that of existing literature studies.
© The Authors, published by EDP Sciences, 2022
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.