Airport operations currently rely on fixed taxi times (default and variable for CDM airports, static for non-CDM airports), without accounting for factors such as weather conditions or ongoing airport work. Additionally, users often receive late notifications regarding stand allocations and the runways in use for departures and arrivals.
This lack of real-time information and dynamic adjustments to taxi times results in overestimated fuel requirements for airlines and inefficient resource planning at airports, including stand allocations and ground handler availability.
The TITOP project was proposed by Swiss and started in January 2023 to address these challenges by providing dynamic taxi-time values for inbound and outbound flights considering weather situations and works at the airports, several hours before departure.
The project combines three machine learning models trained on historical data, which can be used together or individually:
- DE-ICING PREDICTION: this model provides the probability of de-icing for a given flight and whether the de-icing will be remote or at the stand.
- RUNWAY-IN-USE PREDICTION (PRIU): the model predicts which runway will be used for an inbound or outbound flight.
- TAXI-TIME PREDICTION (TITOP): the model provides the prediction of taxi duration between the landing runway and the stand for an inbound flight and between the stand and the departure runway for an outbound flight.
Benefits of these predictions include:
- Operational: More efficient stand planning for airports, better awareness for ground-handlers planning, and more accurate flight plan information
- Economic and Environmental: Reduced and optimised fuel consumption
The project is running a live trial until end of March 2025 to assess the impacts of the three models on operations.
Stakeholders involved:
Airlines: Austrian, Vueling, SunExpress, Swiss, Air France, Transavia France, Transavia Netherlands
Airports: Brussels, Heathrow, Aéroports de Paris, IGA Istanbul, Prague, Vienna, Stockholm