Abstract: This research paper presents a study on the the application of machine learning methods for predicting flight delays. The objective of this research is investigating the ability of different machine learning approaches to forecast flight delays and to identify the most significant factors affecting flight delays. The study is conducted using a comprehensive dataset that includes information on airline schedules, airport congestion, weather conditions, and other relevant factors. The paper begins with a literature review of existing studies on predicting flight delays using machine learning techniques. The performance of Several machine learning techniques, such as decision trees, random forests, support vector machines, and neural networks are assessed and contrasted based on metrics such as accuracy, precision, recall, and F1 score. The outcome of the study exemplifies that machine learning algorithms are highly effective in predicting flight delays, with decision trees and random forests performing the best. The study also identifies weather conditions, airline-specific factors, and airport congestion as the most significant factors affecting flight delays. The inferences from this research paper have significant ramifications for the aviation sector, as precise projection of flight delays can assist airlines and airports better manage their operations and improve passenger satisfaction. Overall, this research demonstrates the capacity of machine learning techniques to improve the accuracy and efficacy of flight delay predictions, which can ultimately lead to a more reliable and efficient aviation system.
Keywords: Flight delays, Machine learning, Data analysis, Feature engineering, Classification algorithms,
Regression algorithms, Decision trees, Feature importance, Performance evaluation, Precision, Recall, F1-score.
|
DOI:
10.17148/IJARCCE.2023.12416
