Air traffic management (ATM) is a sector like no other—where human factors, operational efficiency, and safety imperatives converge in a high-stakes environment. As digitalisation and innovation reshape aviation, the winners of the SESAR Young Scientist Award 2025 are leading the charge, demonstrating how human and machine learning can work in tandem to address the sector’s most pressing challenges.
“During my studies, which initially covered all major transport modes, aviation stood out as the domain in which engineering, human factors, and operational processes are most tightly coupled,” explained Hannes Braßel, Technische Universität Dresden, winner of first prize in the SESAR Young Scientist Award PhD category.
His research looked into enhancing ground safety at airports, with a solution scalable to airports anywhere and to emerging aviation infrastructure.
In aviation, “the interaction between mature, safety-critical systems and the gradual introduction of fundamentally new technologies creates a uniquely demanding engineering environment,” Hannes said.
Improved safety analysis
His winning project combined Light Detection and Ranging (LiDAR) technology, computer vision, and prediction algorithms for aircraft and vehicle movements. Hannes explained that “even small, well-validated improvements in situational awareness or decision support can lead to substantial safety benefits.”
Aircraft ground accidents and incidents were analyzed to identify recurring conflict patterns, which were then recreated in a simulation environment. While such predictive systems may appear sophisticated from the outside, “their foundation is often rather unspectacular,” he added, modestly.
“A significant part of the work consisted of carefully reviewing, classifying, and cross-comparing more than a thousand accident and incident reports, including detailed background research.”
This meant teaming both computer algorithms with human office hours to assess a mass of data. Hannes cautioned that “a central design principle” of his project “is that automation should support human operators rather than replace them.”
The roll out of digitalisation and the evolution of the aviation industry means computers and emerging artificial intelligence (AI) are increasingly playing a role in air traffic control and the way we fly. The European ATM Master Plan 2025 from SESAR maps out this “human–machine teaming,” in which the role of people will evolve significantly, focusing on tasks and situations too complex for humans to handle and teaming up with automation to address emerging traffic challenges.
As Hannes put it: “The aim is not to automate decisions, but to make emerging risks visible at an early stage and in a form that aligns with existing operational reasoning. In this sense, AI is treated as a decision-support layer that enhances situational awareness and enables timely, well-informed human intervention.”
The 2025 jury praised Hannes’ project for being “scientifically excellent, embedding the research within the existing framework and advancing it through innovative developments.”
Tools and the ‘truth’ about airport visibility
The winner in the 2025 Student category agreed that machine learning can best be used to partner rather than replace human expertise.
Vera Tessa Cornelie Buis, Wageningen University & Research in collaboration with LVNL, was given first prize in her category, for research applying machine learning algorithms to improve airport visibility forecasting, especially in low-visibility conditions.
“A machine learning model like this could be a useful additional tool to help forecast visibility conditions,” she explained. “In my opinion, it should be seen as an additional decision-support tool, rather than a ‘truth.’”
Accurate weather forecasts are crucial for minimising delays and optimising flow and capacity management at airports. Specifically, Vera’s work aimed “to understand how machine learning would perform in forecasting a certain weather phenomenon that current weather models struggle with, namely fog.”
She focused on limited runway capacities caused by low visibility conditions. When visibility is low, the number of take offs and landings must be limited for safety reasons. “Sometimes a difference in just one hundred metres can have a big impact on the capacity of a runway,” Vera said.
There is a need to make the aviation industry as efficient as possible, she explained, for reasons ranging from minimising costs and environmental impact, to maximising safety and the passenger experience. The weather plays a significant role in this, directly influencing both on route traffic and airport capacities.
The big advantage of the model used is that “you can tailor it specifically to your liking, which in this case means to the wishes of air traffic control at Amsterdam Airport Schiphol.”
The jury found that Vera’s work showed that “machine learning models, particularly a probabilistic temporal fusion transformer with a custom focal loss function, can reliably predict low-visibility conditions, outperforming traditional numerical weather models.”
For Vera herself, the prize was the culmination of years of dreams and hard work. “I’ve always been into aviation ever since I was young,” she said. “It’s the industry that truly fascinates me the most, combining engineering, hospitality, beauty and hard work in a complex, dynamic and around the clock operation.”
As Andreas Boschen, Executive Director of SESAR JU, said, “The future of air traffic management and aviation will be shaped by work carried out by participants in this year’s awards.”
“Through their research, they help us turn ambitious ideas into tangible innovations that strengthen Europe’s leadership in air traffic management,” He added
Read about past winners of the prize
Read about the 2025 edition of SESAR Innovation Days
