We consider the computation of dynamic arrival routes (DARs): arrival routes from all entry points of a terminal maneuvering area that merge traffic arriving in a given time interval towards the runway for which we can guarantee separation for all arriving aircraft and enable all aircraft to follow optimal descent profiles. We exploit our recent mathematical results to bring the concept of DARs closer to practical applicability by demonstrating how we can integrate both the impact of atmospheric conditions on the aircraft’s descent profiles and the interaction with departing traffic into our DARs framework. With experimental results for Stockholm Arlanda airport, we give a proof of concept of this integration—showing that we can quickly obtain optimal solutions—and highlight the importance of these aircraft-specific and traffic-situation-dependent factors.

We consider two very basic problems – one in unsupervisedand one in supervised learning. In the former, we are given a set ofpoints and have to label half of the points red and half the points blueso as to maximize the red–blue separation, i.e., the length of a shortestbichromatic edge. In the latter, the data (points in the plane) are alreadylabeled red and blue, and we seek a linear classifier (a separator of thetwo given point sets) that can be described using the smallest integers.We give algorithms for both problems. Our solutions are simple; themain contribution of the paper is highlighting the problems and theiralgorithmic solutions, which, to our knowledge, have not been presentedpreviously, despite the problems being fundamental to the field. We alsoconsider related problems.

We present an application of a mixed-integer programming (MIP) framework for automatic traffic synchronization, providing safe separation between the arriving traffic within the terminal maneuvering area (TMA) of an airport implementing point merge (PM) procedures. Additionally, the proposed methodology ensures conflict-free operations when departures and arrivals share a common runway. Based on real traffic scenarios for two European airports, we model realistic descent profiles and assume all the arrivals are performing the most fuel-efficient continuous descent operations (CDOs). We compare two scenarios: in the first, the arriving aircraft are strictly forced to adhere to the published arrival route structures, meaning that a turn towards the merge point may not be initiated prior to reaching the point merge system (PMS), while in the second scenario, aircraft may be assigned a shortcut from a published waypoint along the arrival route. We evaluate the resulting arrival flight efficiency and compare it to that of the actual flights, arriving during the hour selected for our optimization, noticing varying benefits for the two airports and whether shortcuts are allowed or not. Given the correct setting for the specific airport, we demonstrate that our approach provides significant benefits, including increased vertical performance as well as reduced time and distance, contributing to lower levels of noise and fuel savings, accompanied by reduced emissions.

In this paper, we examine the feasibility of assessing air traffic controller (ATCO) workload using non-intrusive eye-tracking measures and machine learning algorithms. A total of N = 18 ATCOs participated in simulator runs with tasks inducing three task-load levels: light, moderate, and heavy. Task load was modulated through traffic load and the associated increase in complexity. We collected eye-tracking data (statistical summaries of which serve as features) and obtained subjective workload assessments using self-reported Cooper-Harper Scale scores, which act as label variables. We evaluate the performance of eight classical machine learning models, with the k-nearest neighbors and support vector classifier models emerging as the most promising. To optimize performance, we apply feature selection techniques, focusing on these best-performing models. Feature selection via recursive feature elimination (RFE) based on permutation importance reduces the original 42 features while maintaining or improving performance. The outcomes yield promising results in workload-level estimation, achieving an F1 score of 0.870 for low/high workload prediction and an F1 score of 0.788 for predicting three different levels of workload. The RFE process identifies optimal feature sets ranging from 7 to 13 features for different tasks, with minimal impact on performance. A "knee point" is observed, representing the optimal balance between model performance and dimensionality. Adding more features beyond this point contributes little to performance improvement while increasing model complexity. These findings indicate that even a few features can be sufficient for accurate workload prediction. We show that head-movement features provide valuable information. Comparable performance is achieved using only ocular features, but this requires more features. Asymmetry in left and right eye metrics holds workload-related information but transforming them into averages and differences reduces performance. Retaining the original features separately is the most effective approach, incorporating their absolute differences may provide slight benefits in certain models.

In this paper, we examine the feasibility of assessing air traffic controller (ATCO) workload (WL) using non-intrusive eye-tracking measures and machine learning (ML) algorithms. We concurrently acquire electroencephalography (EEG) data from a workloadoptimized wearable device and subjective WL assessments through self-reported Cooper-Harper scale (CHS) workload-rating scores, employing both as label variables. A sample of n = 18 ATCOs participate in simulated work sessions encompassing tasks designed to induce three distinct task-load levels: light, moderate, and heavy. We evaluate the performance of five classical ML models. Focusing on the best-performing models, we apply feature selection techniques to identify reduced sets of eye-tracking features. Starting with 58 features, we use a recursive elimination method based on permutation importance, aiming to determine the minimal feature set while also striving for improved performance. The outcomes yielded promising results in the realm of workloadlevel estimation, achieving 96% accuracy (f1-score=0.87) with 34 features for high workload prediction and 88% accuracy (f1-score=0.82) using 57 features in predicting 3 different levels of workload. We further reduced the feature sets to 6-13 features for different tasks with minimal impact on performance. We identified a \x93knee point\x94 as the optimal balance between model performance and dimensionality. Adding more features beyond this point did little to improve performance, but increased model complexity. These results indicate that even a small number (less than 10) of features can be sufficient for WL prediction.

This paper investigates the demand and capacity management in very low level urban airspace, using the notion of Reasonable Time to Act (RTTA) from the European Concept of Operations for U-space (CORUS) and its Urban Air Mobility (UAM) extension (CORUSXUAM). We develop methodology for determining optimal RTTA values which balance early airspace reservation versus the flexibility of last-minute adjustments. Our analysis addresses the trade-offs between early deconfliction (which may lead to inefficient airspace use due to unnecessary reservations), and late deconfliction (which implies reduced predictability for uncrewed aerial systems operators). We incorporate weather uncertainties, using simulations to quantify the impact of varying RTTA lengths on airspace congestion and delay costs.

In this paper, we introduce a Dantzig-Wolfe reformulation to compute aircraft arrival routes in a terminal maneuvering area (TMA) where the aircraft are flying according to theoptimal continuous-descent-operation (CDO) speed profile with idle thrust. This model assumes fixed entry times for all aircraft and a single separation time that is independent of wake-turbulence categories.Preliminary experiments show that this approach leads to significantly reduced runtimes.Moreover, the results indicate the possiblity of extending the reformulation to the full model and applying it to address additional practical considerations, such as wind effects in the future.

Adverse weather is the primary cause of delays to air traffic. In this paper models of different maturity level from the United States, Canada and Europe are compared to derive best practices in how to mitigate these impacts. The models are illustrated through case studies in each one of these airspaces. An example in Jacksonville Center in Florida, one for Toronto Airport and one for the Rhein Airspace adjacent to Munich Airport are presented here. Lastly, some of the modeling characteristics are compared to derive best practices and lesson learned that can be leveraged from each other.