A Malta-led, EU-funded consortium (with Italy, Spain and Switzerland's Skyguide) built an AI tool that predicts dangerous air-traffic congestion up to 60 minutes ahead, versus today's 20-minute window, validated with controllers in simulation but not yet deployed operationally.
60 minutes
Prediction horizon for high-complexity traffic hotspots (validation)
20 minutes
Current operational hotspot warning window (baseline)
1.14 € million
EU financial contribution
Details
Maturity
Pilot
Promoter
University of Malta — Institute of Aerospace Technologies (ASTRA consortium: Deep Blue, INGENAV, Skysoft-ATM, Skyguide)
Period
2023-2026
Region (NUTS)
MT00
Keywords
aviation, air traffic management, transport infrastructure, EU research
Context
ASTRA ('AI-enabled tactical FMP hotspot prediction and resolution') is a research project funded by the EU's SESAR Joint Undertaking, coordinated by the University of Malta's Institute of Aerospace Technologies with partners Deep Blue (Italy), INGENAV (Spain), Skysoft-ATM and the Swiss air navigation service provider Skyguide. It ran from September 2023 to February 2026 with an EU contribution of about €1.14 million.
Objectives
The project targeted a specific operational gap: air-traffic flow-management planners can currently identify a dangerous build-up of aircraft in a given airspace — a 'hotspot' — only around 20 minutes before it occurs, often too little time to reroute traffic efficiently. ASTRA built a machine-learning tool intended to extend that warning window to roughly an hour, while also suggesting resolution strategies to reduce controller workload.
Results
According to SESAR JU's own reporting, the tool achieved 60-minute-ahead prediction of high-complexity 4D traffic areas in validation exercises, assessed through workshops and interviews with operational flow-management experts rather than live-traffic trials. Feedback from these sessions was used to refine the tool's design.
Conclusions
As a completed research and validation project, ASTRA has not yet been deployed into daily air-traffic operations, and no live-traffic accuracy or false-alarm rate has been published; its next step would be operational trials with an air navigation service provider.
Implementation
Indicative cost
Low (< €50k) — EU contribution of about €1.14 million via the SESAR Joint Undertaking, September 2023 to February 2026.
Time to results
Medium (1–3 years) — Ran September 2023 to February 2026; concluded at the research-validation stage without operational deployment.
Staffing & skills
University of Malta Institute of Aerospace Technologies (coordinator), Deep Blue (Italy), INGENAV (Spain), Skysoft-ATM, and Skyguide (Switzerland) consortium partners
Conditions for success
Multi-country consortium including an air navigation service provider (Skyguide)
EU SESAR Joint Undertaking research funding and oversight
Expert validation workshops with operational flow-management planners
Common failure modes
No live-traffic trial or operational deployment yet; validation was qualitative (workshops/interviews) rather than measured accuracy
WFP's AI-assisted HungerMap Live platform, built with Google.org and covering 50+ countries, helped identify worsening hunger in Somalia's Buur Hakaba …
