Bayesian Network Modelling of ATC Complexity Metrics for Future SESAR Demand and Capacity Balance Solutions

Area of Science:

• Air Traffic Management
• Aviation Research
• Complex Systems Analysis

Background:

• Demand and Capacity Management (DCM) are crucial for SESAR (Single European Sky ATM Research) to handle future air traffic growth within Trajectory Based Operations (TBO).
• Current complexity assessment methods inadequately address trajectory uncertainty's impact on air traffic demand predictions.
• This gap hinders the effectiveness of DCM solutions like Dynamic Airspace Configuration (DAC) and Flight Centric Air Traffic Control (FCA).

Purpose of the Study:

• To develop and evaluate Bayesian network (BN) models for quantifying the impact of TBO uncertainties on air traffic demand complexity.
• To analyze the influence of "complexity generators" on "complexity metrics" within specific SESAR 2020 DCM solutions.
• To identify the degree of input variable uncertainty improvement needed to meet desired complexity prediction levels.

Main Methods:

• Development of seven Bayesian network (BN) models tailored to specific DCM concepts (DAC and FCA) and time horizons.
• Elicitation of BN models to represent relationships between trajectory uncertainties and complexity metrics.
• Evaluation of model performance in predicting complexity and identifying key uncertainty drivers.

Main Results:

• The proposed BN models effectively identify and quantify the impact of Trajectory Based Operations (TBO) uncertainties on air traffic demand complexity predictions.
• The models demonstrate the influence of specific "complexity generators" on the chosen "complexity metrics" for DAC and FCA.
• Analysis reveals the necessary improvements in input variable uncertainty to achieve target complexity prediction quality.

Conclusions:

• Bayesian networks offer a robust framework for integrating trajectory uncertainty into air traffic complexity assessment.
• Accurate complexity prediction is vital for the successful implementation of advanced DCM solutions in future air traffic management.
• The developed models provide actionable insights for enhancing the resilience and efficiency of air traffic control systems.