Weining Zhang

Dynamic airspace sectorization (DAS) is a core technology in airspace management, aimed at addressing continuously changing traffic patterns and unpredictable severe weather. Regarding the two fundamental issues of when-to-do and how-to-do a sectorization, a data-driven control framework is developed in this paper, which includes two key modules: the knowledge construction module and the decision execution module. The knowledge construction module mainly defines the knowledge related to airspace sectorization problem, including sector generation model, controller workload model, and sector similarity model. Based on these underlying components, a multi-objective receding horizon optimization (RHO) model is constructed to generate sectorization schemes with gradually changing shapes, solving the task of how-to-do a sectorization. The decision execution module introduces multi-criteria decision analysis and automated decision logic to solve the task of when-to-do a re-sectorization, including selecting an optimal sectorization scheme and analyzing its effectiveness for future traffic situations. Using trajectory data from the Singapore Flight Information Regions (FIRs), a comprehensive assessment and analysis of the two modules is conducted. Empirical findings indicate that, compared with the single interval optimization (SIO) approach, our framework can generate serialized optimal airspace sectorization schemes with high similarity, which will enhance the flexibility of future airspace management.

The optimization of gate allocation and taxiway routing represents a critical challenge in enhancing airport ground operations performance. To simultaneously address these two closely coupled tasks, their interconnected processes are first modeled as flows in a spatiotemporal graph. On this basis, we develop a multi-objective optimization approach that accounts for both temporal and spatial factors across different operational aspects, effectively balancing the diverse needs of travelers, carriers, and airport authorities. To mitigate differences in scale and preference among various optimization objectives, min-max normalization combined with the linear weighting method is employed to transform the multi-objective problem into a single-objective one, which is solved by binary integer linear programming. Based on the actual operational data of Terminal 1 at Shanghai Pudong International Airport, three typical scenarios of different complexity are constructed for validation purposes. Performance comparisons with the state-of-the-art methods demonstrate the superiority of the proposed model in terms of various operational costs and parameter sensitivity. The integrated scheduling solution offers airport operators a reliable and efficient decision-making tool with practical applicability.

To address the issue of missing position in flight trajectory data collected by Automatic Dependent Surveillance-Broadcast (ADS-B) systems, a flight trajectory tensor completion model based on truncated Schatten p-norm minimization is proposed. First, the low-rank characteristics of the trajectory set are validated using Singular Value Decomposition (SVD); based on this, the data is transformed into a three-dimensional tensor structure. Next, a regularization strategy combining the Schatten p-norm with a singular value truncation mechanism is introduced to construct the trajectory tensor completion model, which suppresses noise and interference from minor components while preserving the main variation patterns of the trajectories. Finally, the model is optimized and solved using the Alternating Direction Method of Multipliers (ADMM) to obtain the completed trajectories. Taking historical ADS-B trajectory data from Orly Airport to Toulouse Airport as an example, the completion results of the proposed model under different missing patterns, missing rates, and flight phases are analyzed from both qualitative and quantitative perspectives. Experimental results show that compared with other representative models, the proposed model achieves the best completion performance under different missing patterns and missing rates; the completion performance during the cruise phase is better than during the ascent and descent phases. The proposed model can serve as a preprocessing technique for flight trajectory data in air traffic, providing more complete and reliable data support for various downstream applications.

Controller subjective evaluation is one of the most important ways to assess air traffic complexity. However, the inconsistency of human experts has a negative impact on the inference of complexity analysis models. To solve this problem, this paper proposes to construct a weakly supervised air traffic complexity dataset using highly reliable traffic situation similarity as labelling information. On this basis, a distance metric learning model is trained to generate a distance metric matrix that satisfies the similarity relationship. Finally, the K-means algorithm is combined to realise preferred complexity situation level classification and evolution analysis. Taking the actual operating data of a mid-southern area sector of China as an example, the effectiveness of the proposed method is verified. Experimental results show that the aircraft density, aircraft ground speed variance, heading disorder, convergence speed and horizontal conflict have a greater impact on the complexity situation. Compared with the K-means algorithm based on Euclidean distance, metric learning improves the optimal silhouette coefficient and Davidson-Boldin index by 31.80% and 12.97%, respectively. In addition, it is confirmed that the situation evolution is driven by one or two key influencing factors.

In this paper, a data-driven framework aimed at investigating how weather factors affect the spatio-temporal patterns of air traffic flow in the terminal maneuvering area (TMA) is presented. The framework mainly consists of three core modules, namely, trajectory structure characterization, flow pattern recognition, and association rule mining. To fully characterize trajectory structure, abnormal trajectories and typical operations are sequentially extracted based on a deep autoencoder network with two specially designed loss functions. Then, using these extracted elements as basic components to further construct and cluster per-hour-level descriptions of airspace structure, the spatio-temporal patterns of air traffic flow can be recognized. Finally, the association rule mining technique is applied to find sets of weather factors that often appear together with each flow pattern. Experimental analysis is demonstrated on two months of arrival flight trajectories at Hong Kong International Airport (HKIA). The results clearly show that the proposed framework effectively captures spatial anomalies, fine-grained trajectory structures, and representative flow patterns. More importantly, it also reveals that those flow patterns with non-conforming behaviors result from complex interactions of various weather factors. The findings provide valuable insights into the causal relationships between weather factors and changes in flow patterns, greatly enhancing the situational awareness of TMA.