Paolino De Falco
De Falco is a project leader and data scientist at Eurocontrol, leading innovative projects, including PRELUDE and OpTT2.0, focusing on predictive tools for luggage delivery times and turnaround time across European airports. De Falco is skilled in applying machine learning to improve operational efficiency, with a background in academic research and consulting and a proven track record of driving collaboration with industry stakeholders, including airlines and airports, bringing AI-driven solutions into real-world applications. De Falco is passionate about leveraging data to enhance decision-making in air traffic management.
Day -
1 - Management and operations – increasing capacity and efficiency, airside - 10:55 - 11:20
Machine learning automation of target off-block time
Airports utilizing collaborative decision-making (A-CDM) rely on accurate estimates of target off-block time (TOBT) set by airlines or ground handlers. This time indicates the end of ground handling, including possible ground delays, and should be updated whenever changes in the handling progress occur. Last-minute changes in TOBT can disrupt pre-departure sequence and air traffic flow management, making its timely and accurate estimate vital. As one of the EATIN (Eurocontrol Air Transport Innovation Network) initiatives, this work presents a machine learning model for TOBT predictions, their implementation into Prague Airport’s operating system and their distribution to stakeholders and the network manager.
The audience will learn:
- A machine learning model for turnaround and target off-block time predictions
- The validation steps and integration of machine learning into Prague Airport's operating system
- The approach used at Prague Airport to automatically distribute the model's outcome to stakeholders and the network manager
- An analysis showing the acceptance of the model's outcome by ground handlers at Prague Airport
