This project is conducted with using flight delay data from Delta Airline at JFK between 2016-17.
In this project, I am assuming myself as a business strategy analyst working for Delta Air Lines. This is a flight delay analysis for all the Delta US domestic flights departing and arriving at John F. Kennedy Airport in New York. There are two parts of my analysis. First, I would compare the relationship between the features and the delays to understand the main causes of the delays and determine what can be used as training features This is also the data cleaning before training the models. Main dataset of flight delays data for the period 2016/11 – 2017/ 10 was obtained from the Bureau of Transportation Statistics (BTS) which is a database maintained by the U.S. Department of Transportation. Secondly, I would establish three classification models and use the preprocessed data to predict flight delays. After comparing the results from all models, I recommend Random Forest Classifier which provides predictions most accurately and efficiently. This information is expected to assist airline management in flight planning and delay remediation procedures. Effectively minimizing the damage of delayed flights can optimize the airline’s overall profits.
