A machine learning approach for predicting airport passenger flows

The European aviation policy envisages a resilient air transport system seamlessly integrated into the European transport network, with the final objective of taking passengers from door to door predictably and efficiently while enhancing air transport experience. Meeting this vision calls for an in-depth understanding of air passengers’ travel behavior and the ability to anticipate its impact on the performance of the air transport system. This paper presents a methodology to forecast the upcoming airport passenger flows for a particular day of operations, in order to help airports make more informed passenger flow management decisions and render the air transport system more resilient against adverse circumstances.