The Federal Aviation Administration (FAA) announced on November 10 that it will deploy its SMART — Strategic Management of Airspace, Routes and Trajectories — system across all 22 air route traffic control centers nationwide, extending the reach of its AI-powered platform designed to predict and mitigate congestion in the U.S. national airspace system.
The tool, which began operations in the Washington, D.C. region on September 21, has reportedly rerouted 180 flights, saving an average of 10 minutes and nearly 80 gallons of fuel per diverted flight, according to FAA figures. The system integrates approximately 200 data streams — weather conditions, flight paths, traffic volumes and controller staffing metrics — to identify potential bottlenecks before they materialize and suggest schedule or route adjustments.
SMART's architecture includes a chain of human validation: FAA specialists in Washington analyze the AI's recommendations and forward them to the Air Traffic Control System Command Center in Warrenton, Virginia, which distributes them to individual facilities. It is local managers who decide whether to implement the suggestions, and air traffic controllers retain ultimate responsibility for safe aircraft separation. The FAA reiterates that SMART will never take direct control of an aircraft or replace human controllers.
The project is part of a broader $875 million, 12-year contract awarded to Air Space Intelligence in June, which also includes the Flow Management Data and Services (FMDS) platform described as the future technology backbone of the Command Center.
The industry's reaction is mixed. Executives from major airlines such as Delta and American Airlines issued statements of support for the expansion, citing tangible operational benefits. Ed Bastian, Delta's CEO, and Robert Isom of American Airlines were cited by the FAA as supporters of the system from its initial regional launch.
Conversely, the National Air Traffic Controllers Association (NATCA), the union representing more than 15,000 controllers, told Flying magazine that it was never consulted on the design, testing, or implementation of SMART. The union warned it cannot assess the system's real impact on controllers' workload and safety — a concern that echoes a pattern already observed in other AI deployments in critical infrastructure, where exclusion of operational professionals breeds resistance and mistrust.
Those concerns carry extra weight given that more than two dozen members of Congress sent a letter on October 9 calling on the FAA to suspend the initial testing phase until adequate answers were provided on safety assessments, personnel training and oversight mechanisms. The November 10 announcement did not address these questions.
SMART operates against a backdrop of growing pressure on the U.S. civil aviation system. Congestion in the airspace — worsened by cascading delays, severe weather and sustained post-pandemic traffic recovery — has cost billions in wasted fuel and lost productivity. According to AvWeb, the system was first deployed in the airspace surrounding Reagan National, Washington Dulles and Baltimore/Washington International airports.
For sector analysts, the expansion of SMART marks a turning point: for the first time, an AI-based machine learning system is being deployed at national scale to manage one of the world's most heavily regulated and critical sectors. What happens next — whether the system actually delivers on its promises, whether controller resistance intensifies, whether Congress demands audits — could serve as a template (or a cautionary tale) for other nations considering similar solutions for their own airspace.
Sources: AvWeb, AeroTime, Atlabyte
✓ Independent sources cross-checked and verified before publishing
