Flight Delay Analysis and Prediction Using Machine Learning Algorithms

Abstract

The demand for air transport has increased significantly with the rapid development of the global economy. The flight delays have been observed as one of the toughest problems in aviation sector. Flight delays are not only inconvenient for customers, but they also cost airlines income. Accurate flight delay estimation is crucial for airlines since the data can be used to improve passenger service and airline agency revenue. In this paper proposed model is designed in such a way that can predict departure delays, and another model that can classify arrival delays. In this project author apply machine learning algorithms using Linear Regression, Random aSampling, and Polynomial Regression. The purpose of proposed model is not to obtain the best possible prediction but rather to emphasize on the various steps needed to build such a model. Experiments based on realistic datasets obtained from Kaggle of Flight. The experimental analysis achieve a test accuracy of approximately 72%.