Integrating Safety Metrics into the Arrival Capacity Management of large Aerodromes

Project information

Funding: DFG - German Research Foundation
Duration: 36 month

The airspace/airport capacity is critical for the current approach control by Air Traffic Control (ATC) in a Terminal Maneuvering Area (TMA) at complex airports. Automation tools are required to support complex and heavily constrained ATC operations, in order to achieve more efficient approach traffic and reduce the workload of Air Traffic Controllers (ATCOs). This project aims to design optimization/simulation process necessary for the automation, model such capacity-critical approach control, and integrate it into the optimization/simulation. This is carried out by utilizing the simulation environment relying on Agent-Based Modelling and Simulation (ABMS) for safety analysis developed earlier by the Chair of Air Transport Technology and Logistics of TU Dresden. In this project, special attention is paid to the aspect of safety.

Today's approach-control procedures are based on a safety assessment regarding route design, procedural rules and boundary conditions that is carried out upstream in the descent/approach process. However, the approach operation is actually influenced by significant fluctuations of weather, traffic composition etc. Thus, only operational performance indicators are recorded and reported in accordance with SES Performance Scheme. This implies the safety is deemed implicitly assured. ATC deals with this factual error by applying conservative and capacity-sensitive inter-aircraft separation rules to approaching aircraft. However, this does not lead the system permanently to an overall optimal operation.

Integrating the safety aspects into the aforementioned performance assessment as an online process will enable safety analyses to be constantly incorporated into performance assessment on a tactical level. This will eventually maximize additional airspace and takeoff-landing runway capacity. To achieve this, the correlation between quantitative safety metrics (collision/incident probabilities) and various ecological/economic performance indicators (capacity, environment, cost efficiency) will be studied. A positive correlation is postulated between hazard potential and so-called "traffic complexity" as entropy, i.e., as a measure of disorder or a measure of workload in the traffic system. As these two factors obviously conflict with each other, we will carry out a trade-off study. To investigate all of these research aims, we apply multi-objective optimization methods to our agent-based fast-time simulation.

We aim to design the safety/performance-analysis model so that it can prognose future hazardous events. We employ machine learning methods and system modeling techniques to achieve the necessary foresight into the upcoming traffic situation. We develop a decision support system for an approach traffic, which is dynamically adaptive to the changing situation. All these analyses and developments will finally lead to an increase in the operational efficiency. The system is validated for some German airports.

Collaborators are: DFS (Deutsche Flugsicherung GmbH), Munich airport, and Frankfurt airport. They will ensure the actual implementation.

This project is divided into the following six work packages (WPs):

WP 1: Research on ATM performance indicators, their correlations, and relevant methods
WP 2: Determination of “safety” and “capacity” in highly automated ATM systems
WP 3: Modeling of hazards and countermeasures in approach
WP 4: Implementation of a Safety-driven Arrival Capacity Manager
WP 5: Analyses on robustness of safety-capacity correlations
WP 6: Evaluation of results and gaining insights