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

Reducing aircraft taxiing emissions will deliver a significant contribution to the worldwide goal of net-zero greenhouse gas emissions in the aviation industry. Replacing jet-engine taxiing by towing aircraft with electric towing vehicles is expected to reduce taxiing emissions by roughly 80%. Introducing a fleet of towing vehicles introduces operational challenges to an airport. Although there has been research focused on optimizing the assignment of vehicles to aircraft, such an assignment will require changes during a day of operations, when disruptions such as flight delays occur. This paper proposes two models, a strategic and a disrupted model, with which an adaptive vehicle-to-aircraft assignment is created. The models are formulated as Mixed Integer Linear Problems, and both maximize the number of towed aircraft and minimize the schedule changes for vehicle operators. The approach illustrated includes vehicle and aircraft routing, conflict avoidance, and a model for energy usage. We apply the models to Amsterdam Airport Schiphol, where the disrupted model is able to create assignments that remain the same in subsequent time steps for an average of 55% of the vehicles, on a busy day, when towing all aircraft. Furthermore, the results show that minimizing schedule changes does not come at the expense of fewer towed aircraft, i.e. of smaller emission savings. Lastly, we investigate the impact of fleet size and general on-time performance on the assignments created by the model.

A key part of efficient airport operational planning is to have insight into potential flight delays and cancellations. For airport planners, it is important to obtain flight delay or cancellation predictions with a high degree of certainty, i.e. a high precision. This allows planners to make sound decisions based on these predictions. To obtain such predictions, machine learning classification techniques are often applied. An important issue for classification problems is that of imbalanced class distributions: the number of actually cancelled/delayed flights is low. In general, the imbalance is addressed by resampling the data using one or more sampling techniques. However, resampling does not necessarily correspond to an imbalance ratio that leads to the best classification results. In this paper a systematic approach is presented to deal with imbalanced data for classification problems, while taking into account the preferences of airport planners. A range of feasible imbalance ratios, together with several classification algorithms and sampling techniques, are considered. An optimal imbalance ratio is identified with respect to relevant performance metrics. The approach is illustrated by performing binary classification of flight cancellations and delays at a large European airport. The results show that the highest prediction precision is obtained using a base imbalance ratio, whereas a higher imbalance ratio is needed to obtain the highest F1-score. Specifically, the cancellation prediction performance is increased by up to 243%, while its optimal imbalance ratio does not correspond to resampling. In general, the results underline the need to investigate the influence of varying data imbalance ratios on the performance of classification algorithms.