The EUROCONTROL Network Manager is currently trialling the use of artificial intelligence (AI) to predict the impact of likely regulations at large airports in Europe. Making use of the most up-to-date technologies, the project "Regulation Outlook" – launched under SESAR 2 (PJ04, solution 28.3) is displayed pre-tactically and tactically in the Network Manager’s Airport Function position, providing not only the probability that an Airport regulation will be needed , but also explanations about the most important parameters that determine the presence of a regulation, whether they are industrial action, weather, air traffic control or aerodrome capacity issues. Currently, 'Regulation Outlook' estimates the likelihood of a regulation being required at any of the network’s busiest airports with up to 87% accuracy – which is already helping NM coordinate with operational actors to anticipate better any possible disruptions flagged by the new tool.
Predicting the impact of likely regulations at large airports in Europe: EUROCONTROL NM trials the use of AI
EUROCONTROL’s Network Manager Operations Centre uses AI to predict the impact of likely regulations at large airports in Europe
“The EUROCONTROL Network Manager is using artificial intelligence and machine learning to better predict capacity reductions that are likely to affect Europe’s largest airports and in consequence the air traffic flow throughout the entire Network. State-of-the-art applications such as this enable us to check and coordinate immediately with all actors to agree on the most appropriate measures that will minimise delay.”
Initially, the predictions of capacity reductions at airports were based on information available the day before operations (D-1). NOTAMs, weather predictions, traffic predictions and the EUROCONTROL NM’s Airport Corner’s events were the sources to be used to determine whether a regulation at an airport is likely to happen. The first phase of 'Regulation Outlook' was aimed at building a minimum viable product to check the feasibility of regulation predictions with a few input features: METARs, Demand Data Repository consolidated flight intentions (CFIs), and basic NOTAM information extracted from the Q-code field. The output would be expressed in terms of the probability that a regulation could happen in one of the top 100 airports in the network. As the results proved satisfactory, more functionalities were incorporated into the predictive model: Terminal Aerodrome Forecasts (TAFs) including longer-term weather predictions, forecasted traffic count predictions based on CFIs, and the addition of Airport Corner events information.
Next steps: Once the trial has finished, the plan is to share daily predictions with operational stakeholders and extend the look-ahead time up to 6 days.
Latest highlights
First in Europe: Türkiye’s DHMI implements triple runway operations at iGA Istanbul Airport
