Skip to content
AI.info

The Pulse

FAA Sets September 21 AI Trial for Washington Airspace

The FAA is preparing to deploy SMART, an AI-assisted air-traffic planning system, around Washington, D.C., under a 12-year, $875 million contract with Air Space Intelligence. The platform will analyze schedules, weather, airport capacity an

FAA Sets September 21 AI Trial for Washington Airspace

AI.info Team ·

SMART is due to enter service near Washington

The Federal Aviation Administration is preparing to put an AI-assisted air-traffic planning system into operation around Washington, D.C., as soon as Monday, September 21. The initial deployment will cover the region’s three major airports and serve as the first operational test of a system the agency plans to expand across the national airspace.

SMART, short for Strategic Management of Airspace, Routes, and Trajectories, is part of a 12-year, $875 million contract awarded to Boston-based Air Space Intelligence in June. The company is also developing Flow Management Data & Services, a related system intended to replace the existing platform used at the FAA’s Air Traffic Control System Command Center in Virginia.

The FAA’s SMART fact sheet describes the software as a cloud-based planning layer that works alongside existing air-traffic systems. It will not take over the job of separating aircraft: the agency says controllers will remain responsible for keeping flights safely apart.

What the $875 million system will analyze

SMART is designed to give the FAA, airlines and other operators a shared view of conditions before flights depart. Its AI will analyze airline schedules, weather, airport capacity, airspace conditions and other operational constraints to predict traffic flows and identify possible conflicts before they develop.

The system will continuously update its forecasts and recommendations as conditions change. The FAA says those recommendations can help operators agree on routes and departure times, reduce fuel consumption, improve on-time performance and speed recovery after weather or congestion events.

Air Space Intelligence said in its June 22 contract announcement that the award covers both predictive modeling and decision-support software. FAA Administrator Bryan Bedford said the systems would help controllers “better anticipate demand, balance capacity, and manage traffic before delays occur.”

“FMDS and SMART will give controllers modern, data-driven tools to better anticipate demand, balance capacity, and manage traffic before delays occur.”

Bryan Bedford, FAA administrator

Airlines want limits stated clearly

The planned launch follows weeks of uncertainty among airlines about how the software would affect operations. The FAA has told the industry that SMART will not change existing procedures for controllers or airlines. Instead, it will produce alternative route information through FAA systems already used by operators.

The distinction matters because the software is meant to advise planners before flights depart, not issue autonomous commands to aircraft in the air. The FAA fact sheet presents SMART as a system for identifying constraints earlier, when planners still have time to adjust routes, schedules and traffic flows.

Philip Mann, principal consultant at Vector Strategic Consulting and a former FAA employee, told Ars Technica that a limited first deployment reduces the number of unknowns in a system that could eventually operate across the U.S. airspace network. He said the key tests will involve performance under real workloads and degraded data, along with clear responsibility when a prediction is wrong.

The trial arrives during a much larger FAA rebuild

SMART is one component of a broader FAA effort to replace aging air-traffic infrastructure. A September 2026 Government Accountability Office report says the agency’s modernization program includes communications, surveillance, weather and training systems in its first phase, while a second phase is expected to develop new automation for tracking aircraft and optimizing traffic.

The GAO says the first phase is projected to cost $10.6 billion, compared with $9.03 billion in identified appropriations. The report also says the FAA has not produced a comprehensive lifecycle cost estimate or an integrated master schedule for the program. More than 11,000 individual project schedules remained unintegrated as of May 2026.

Those findings put the $875 million AI contract inside a modernization effort whose technical ambition is matched by management risk. The agency wants SMART and FMDS to provide a modern planning layer while other contractors replace radar, communications and display systems that support the same national network.

Human control remains the test

Air Space Intelligence says its Flyways AI platform already supports flight routing and traffic-management work for commercial and government customers. The FAA contract gives the company a much larger assignment: supplying software at the center of the agency’s effort to reduce delays and manage demand across the National Airspace System.

The first test will be narrower than that eventual goal. SMART is expected to begin with Washington-area operations, where dense traffic, weather disruptions and constrained airport capacity provide a demanding environment without immediately exposing the entire system to a nationwide rollout.

The FAA’s stated measure is not whether SMART can produce a forecast, but whether its forecasts help people make better decisions before congestion spreads. On September 21, the agency is scheduled to begin testing that proposition with controllers and operators still responsible for the final calls.

Source

The Wall Street Journal

Explore

More articles