Search Results (3)

The rapid growth in air traffic has led to increasing congestion at airports, creating bottlenecks that disrupt ground operations and compromise the efficiency of air traffic management (ATM). Ensuring the predictability of ground operations is vital for maintaining the sustainability of the ATM sector. Flight efficiency is closely tied to adherence to assigned airport arrival and departure slots, which helps minimize primary delays and prevents cascading reactionary delays. Significant deviations from scheduled arrival times—whether early or late—negatively impact airport operations and air traffic flow, often requiring the imposition of Air Traffic Flow Management (ATFM) regulations to accommodate demand fluctuations. This study leverages a data-driven machine learning approach to enhance the predictability of in-block and landing times. A Bidirectional Long Short-Term Memory (BiLSTM) neural network was trained using a dataset that integrates flight trajectories, meteorological conditions, and airport operations data. The model demonstrated high accuracy in predicting landing time deviations, achieving a Root-Mean-Square Error (RMSE) of 8.71 min and showing consistent performance across various long-haul flight profiles. In contrast, in-block time predictions exhibited greater variability, influenced by limited data on ground-level factors such as taxi-in delays and gate availability. The results highlight the potential of deep learning models to optimize airport resource allocation and improve operational planning. By accurately predicting landing times, this approach supports enhanced runway management and the better alignment of ground handling resources, reducing delays and increasing efficiency in high-traffic airport environments. These findings provide a foundation for developing predictive systems that improve airport operations and air traffic management, with benefits extending to both short- and long-haul flight operations. Full article

Meeting the demand with the available airspace capacity is one of the most challenging problems faced by Air Traffic Management. Nowadays, this collaborative Demand–Capacity Balancing process often ends up enforcing Air Traffic Flow Management regulations when capacity cannot be adjusted. This process to decide if a regulation is needed is time consuming and relies heavily on human knowledge. This article studies three different Air Traffic Management frameworks aiming to improve the cost-efficiency for Flow Manager Positions and Network Manager operators when facing the detection of regulations. For this purpose, two already tested Deep Learning models are combined, creating different hybrid models. A Recurrent Neural Network is used to process scalar variables to extract the overall airspace characteristics, and a Convolutional Neural Network is used to process artificial images exhibiting the specific airspace configuration. The models are validated using historical data from two of the most regulated European regions, resulting in a novel framework that could be used across Air Traffic Control centers. For the best hybrid model, using a cascade architecture, an average accuracy of 88.45% is obtained, with an average recall of 92.16%, and an average precision of 86.85%, across different traffic volumes. Moreover, two different techniques for model explainability are used to provide a theoretical understanding of its behavior and understand the reasons behind the predictions. Full article

The European Air Traffic Management Network (EATMN) is comprised of various stakeholders and actors. Accordingly, the operations within EATMN are planned up to six months ahead of target date (tactical phase). However, stochastic events and the built-in operational flexibility (robustness), along with other factors, result in demand and capacity imbalances that lead to delayed flights. The size of the EATMN and its complexity challenge the prediction of the total network delay using analytical methods or optimization approaches. We face this challenge by proposing a deep convolutional neural network (DCNN), which takes capacity regulations as the input. DCNN architecture successfully improves the prediction results by 50 percent (compared to random forest as the baseline model). In fact, the trained model on 2016 and 2017 data is able to predict 2018 with a mean absolute percentage error of 22% and 14% for the delay and delayed traffic, respectively. This study presents a method to provide more accurate situational awareness, which is a must for the topic of network resiliency. Full article