Probabilistic Flight Delay Predictions Using Machine Learning and Applications to the Flight-to-Gate Assignment Problem

The problem of flight delay prediction is approached most often by predicting a class or value. However, the aviation industry can benefit greatly from probabilistic predictions on an individual basis, as these give insight into uncertainty predictions. Therefore, in this study, two forecasting algorithms, Mixture Density Networks and Random Forest regression, are applied to predict delays at European airport. algorithms estimate well distribution arrival departure with Mean Absolute Error less than 15 min. To illustrate utility estimated distributions, we integrate flight-to-gate assignment problem. objective increase robustness assignments. Considering predictions, our proposed model reduces number conflicted aircraft up 74% when compared deterministic model. In general, results considering for robust airport operations’ optimization.