FAA tees up $875M AI tool to help manage air traffic congestion
FAA's $875M, 12-year contract with Air Space Intelligence will deploy the SMART AI system to predict and manage US air traffic congestion.
Air Space Intelligence won an $875 million, 12-year FAA contract in June covering the SMART AI tool and a new Flow Management Data and Services system to replace existing infrastructure at the Air Traffic Control System Command Center in Virginia. The company claims its Flyways AI platform, built on a 4D digital twin of US national airspace, already helps manage over 40% of US air traffic through customers like Alaska Airlines. It is unclear which AI model types SMART uses, and experts are watching whether rollout beyond aircraft at 24,000 feet and above will be gated by real-world performance and clear accountability for wrong predictions.
- $875M, 12-year FAA contract awarded to Air Space Intelligence in June
- SMART serves as predictive layer over the new Flow Management Data and Services backbone
- Flyways AI platform reportedly manages over 40% of US air traffic via Alaska Airlines and others
- Unclear whether SMART uses deterministic, machine-learning, or generative AI models
- Initial rollout targets aircraft at 24,000 feet and above amid FAA radar and radio upgrades
Full article374 words · extracted from arstechnica.com · click to collapse
Pinning down the AI
SMART is part of an $875 million, 12-year contract awarded to the Boston-based company Air Space Intelligence in June. The contract also covers development of a Flow Management Data and Services system meant to replace the current system in the FAA’s Air Traffic Control System Command Center in Virginia.
In an article for Global Airspace Radar Magazine, Mann described the new Flow Management Data and Services system as the “backbone,” whereas SMART represents the “predictive layer above it.”
Air Space Intelligence claims its separate Flyways AI platform already helps manage over 40 percent of all US air traffic through partnerships with customers like Alaska Airlines. That platform incorporates a 4D digital twin of US national airspace to help predict air traffic and weather conditions.
It is unclear what types of AI models are being used in the new SMART system. Older deterministic AI models follow predefined rules and logic to consistently produce the same output every time, whereas machine-learning models can statistically analyze patterns to make predictions based on the probabilities at hand.
By comparison, generative AI models—such as the large language models powering popular AI chatbots—generate outputs that can change every time and may also be inaccurate.
Ars reached out to the FAA for comment.
Mann wants to know whether the SMART rollout is still limited to managing aircraft flying at 24,000 feet and above, as the FAA described in June. In that case, the system’s software would be handling “cruise traffic transiting that airspace, plus the climbs and descents feeding the metro airports,” Mann explained.
“What I will be watching is what evidence gates the next expansion: performance under real workload and degraded data rather than demonstration conditions, and a written answer to who owns the outcome when a prediction is wrong,” Mann told Ars.
Under pressure
The government’s plan to launch AI-assisted air traffic management comes as the FAA manages a multi-billion-dollar effort to upgrade aging equipment that supports air traffic control and management. The agency has just started replacing hundreds of radar systems dating back to the 1980s, along with radios, telecommunication lines, and the voice switches that connect calls between air traffic controllers and pilots, Mann pointed out in a Substack post.