Key Takeaway
The U.S. Federal Aviation Administration is beginning a phased rollout of an AI-enabled air-traffic planning system designed to identify congestion before aircraft leave the ground. The model does not hand safety decisions to an algorithm; it adds a predictive planning layer around human controllers. For Georgia, copying the U.S. system would make little sense, but the underlying shift-from reacting to disruption toward anticipating it-matters as passenger traffic, flight activity and airspace use continue to grow.
Analytical Introduction
Airport operations can look linear from the outside: an aircraft lands, passengers move through the terminal, and another flight departs. In practice, every movement depends on a network of interlocking decisions-departure timing, routing, sector capacity, weather, runway availability and the plans of other airlines. As traffic grows, a delay in one part of the network can propagate through many others.
The U.S. Federal Aviation Administration is trying to address this problem through Strategic Management of Airspace, Routes, and Trajectories, or SMART. The September 18, 2026 edition of The Wall Street Journal reported that the tool was preparing to launch in the Washington, D.C. region before a broader rollout. It is designed to combine airline schedules, weather, runway capacity and operational constraints so that potential conflicts can be identified before they become actual disruptions.
FAA materials are explicit that SMART is not a replacement for air-traffic controllers. It acts as an additional planning layer and provides a shared, real-time picture of the system. Controllers remain responsible for safely separating aircraft. This distinction is crucial: in aviation, the near-term value of AI is not autonomous substitution for human judgment, but earlier and better decisions from complex operational data.
From Reaction to Anticipation
Traditional air-traffic management often has to respond to conditions as they emerge. Weather deteriorates, airport capacity drops, or a corridor becomes overloaded, and the network then manages the consequences. A predictive model shifts part of that decision earlier-adjusting departure times, routes or trajectory points before congestion materializes.
According to the FAA’s 2026 SMART fact sheet, the system is intended to build a single operational picture from schedules, weather, airport capacity and airspace conditions, then provide predictive insights on demand and capacity. The FAA also identifies potential benefits including reduced fuel burn, better on-time performance and faster recovery from weather or congestion.
None of those gains is automatic. A predictive system is only as good as its data and its integration with human workflows. In a safety-critical sector, algorithmic recommendations need to be transparent, testable and consistent with existing safety procedures. Deployment therefore requires far more than software procurement: it involves data standards, process redesign, staff training and a clear allocation of responsibility.
Why This Matters for Georgia
Georgia’s aviation market is radically smaller than that of the United States, but the core operational challenge-managing growing traffic efficiently-is increasingly relevant. According to Georgia’s Civil Aviation Agency, the country’s airports handled 1,605,500 passengers in January–March 2026, 4.28% more than in the same period a year earlier. A total of 7,799 flights were operated, up 8.70% year on year, while overall use of Georgian airspace increased by 14.19%.
Tbilisi International Airport handled 1,099,204 passengers in the first quarter, an annual increase of 8.02%. Kutaisi International Airport handled 386,238 passengers, up 6.86%. These figures do not imply that Georgian airports face the same congestion pressures as the world’s largest hubs, but they do show that planning quality, airspace capacity and disruption management are becoming more economically valuable.
Georgia also has an important institutional advantage: air-traffic flow management is already integrated into the wider European network. Georgia’s Aeronautical Information Publication states that centralized Air Traffic Flow Management is provided through EUROCONTROL’s Network Manager, with a Flow Management Position at Tbilisi ACC. Georgia therefore does not start from zero; it already has a coordination and data infrastructure on which more predictive decision-support tools could eventually be layered.
Kutaisi as a Future Test Case
Kutaisi International Airport offers the clearest Georgian case study. In July 2026, Georgian Air Navigation Services said construction of the airport’s new 3,600-metre runway was nearing completion, with the airport expected to be capable of handling the world’s largest aircraft from the end of the year. Navigation and airfield-lighting infrastructure is being modernized at the same time, including a new Instrument Landing System, meteorological stations, updated aeronautical data and revised flight procedures.
Expanding infrastructure does not only mean being able to accommodate larger aircraft. As technical capacity and flight volumes grow, the value of coordination across schedules, airspace, ground operations and weather increases. This is where algorithmic planning can be useful: not as a path to an “unmanned airport,” but as a way to prepare human decisions with better information.
A more realistic strategy for Georgia would be to begin with narrow and measurable use cases-for example, early delay-risk detection, comparison of operational scenarios or predictive analytics for traffic-flow planning. Such an approach is less risky than replacing an entire system at once and makes it possible to validate technology gradually in live operations.
Where the Main Risks Are
The largest risks in algorithmic aviation management are not simply “bad AI.” More practical risks include poor data quality, weak interoperability between systems, overreliance on automated recommendations, cybersecurity vulnerabilities and unclear responsibility for final decisions. In a safety-critical environment, an algorithm cannot become a black box whose output is accepted merely because a computer produced it.
Scale is another limitation. The complexity of U.S. airspace and the size of its airline network cannot be directly mapped onto Georgia. The $875 million U.S. contract is therefore not a useful benchmark for what Georgia should spend. The more relevant question is which function produces the most local value: reducing delays, improving routing, managing seasonal peaks, anticipating weather disruption or integrating operational data into one decision environment.
BTU Researchers’ Assessment
According to an assessment by BTU researchers, the most realistic path for Georgian aviation is not to replace controllers with algorithms, but to add predictive decision support gradually on top of the existing European flow-management environment. As passenger traffic, airspace use and infrastructure capacity expand, particularly around Kutaisi, the economic value of such tools is likely to increase. Their effectiveness, however, will depend on four conditions: high-quality data, interoperability with international systems, trained personnel and clearly defined safety accountability.
Georgia’s smaller scale is not only a constraint. It can also be an advantage because new tools can be tested in a more contained operational environment. If early projects are tied to specific operational problems and measurable outcomes, Georgia can avoid expensive “technology for technology’s sake” and concentrate investment on tools that actually improve punctuality, capacity and resilience.
Conclusion
Managing airports with algorithms does not yet mean that computers independently control air traffic. The more realistic international model is AI as a planning and decision-support layer: a system that processes large volumes of operational information and helps humans identify problems before they become real disruption.
For Georgia, the key question is therefore not when airports will be “handed over to AI.” It is where predictive analytics can already improve planning while final safety control remains human. With rising flight activity, integration into the European air-traffic network and the expansion of Kutaisi Airport, this is becoming an infrastructure and management question rather than a distant technology scenario.
Data and Main Sources
- The Wall Street Journal – “FAA to Deploy AI Tool to Help Manage Air Traffic,” September 18, 2026 (user-provided issue)
https://www.wsj.com/ - Federal Aviation Administration – “MODERN SKIES: … Selects Air Space Intelligence to Deploy State-of-the-Art Air Traffic Control Software”
https://www.faa.gov/newsroom/modern-skies-trumps-transportation-secretary-sean-duffy-selects-air-space-intelligence - Federal Aviation Administration – SMART One Page Fact Sheet
https://www.faa.gov/newsroom/SMART_One-Pager.pdf - Civil Aviation Agency of Georgia – 2026 Q1 passenger traffic and flight activity
https://gcaa.ge/2026-%E1%83%AC%E1%83%9A%E1%83%98%E1%83%A1-%E1%83%9E%E1%83%98%E1%83%A0%E1%83%95%E1%83%94%E1%83%9A-%E1%83%99%E1%83%95%E1%83%90%E1%83%A0%E1%83%A2%E1%83%90%E1%83%9A%E1%83%A8%E1%83%98-%E1%83%9B%E1%83%92/ - Sakaeronavigatsia / AIP Georgia – Air traffic flow management and airspace management
https://airnav.ge/eaip/2026-04-16-000000/pdf/UG-ENR-1.9.pdf - Sakaeronavigatsia – Navigation infrastructure modernization for the new runway at Kutaisi International Airport
https://imaps.airnav.ge/en/akhali-ambebi/saqaeronavigatsia-qutaisis-aeroportis-asafren-dasafren-zolze-sanavigatsio-infrastruqturas-srulad-ganaakhlebs
Prepared by the academic team of Business and Technology University and the BTUAI Research Team, Tbilisi, Georgia.