Using Machine Learning to Predict Additional Taxi-Out Time as a Airport Key Performance Indicator in the Eurocontrol Zone

The need to expand the air transport system capacity has increased rapidly in the last few years. European airports are located in urban areas and have limited opportunities for expansion. In addition, there is high pressure on air traffic and is associated with reducing the negative impact on the environment. Therefore, it is important to optimize and transform the system from what already exists into a more efficient and greener one. This paper will evaluate how to use machine learning methods to predict a key performance indicator for airside operation, taxi-out time. Statistical correlation methods will be used to establish the dependence between attributes and the output variable. This will help decision-makers with airport optimization measures and ease the decision-making process at the macroscale level.