FAA starts SMART air traffic rollout in Washington
The FAA has started limited use of SMART around Washington, D.C., turning a June Air Space Intelligence contract into a live test of predictive airspace management.
The FAA has started using SMART in limited mode around Washington, D.C., giving traffic-flow specialists a predictive view of congestion, weather and capacity while keeping aircraft control and facility decisions with humans.
The September 21 FAA announcement says Strategic Management of Airspace, Routes and Trajectories, or SMART, centralizes about 200 data streams, including weather patterns, flight paths, traffic flow and controller staffing metrics. The agency says the tool then gives aviation specialists a way to see future conditions and recommend fixes before problems build.
The first deployment is narrow. The FAA says it began using SMART in the airspace around Washington, D.C. and plans to expand gradually. FedScoop reported that the National Capital Region launch covers Baltimore/Washington International Thurgood Marshall Airport, Dulles International Airport and Ronald Reagan Washington National Airport.
What SMART does now
SMART is a planning and traffic-flow tool, not an automated air traffic controller. The FAA says its recommendations are reviewed by FAA staff, and local leadership at air traffic control facilities can accept or pass on scheduling suggestions.
| Part of the rollout | What the source record supports | Important limit |
|---|---|---|
| Inputs | Weather, flight paths, traffic flow, staffing metrics, airline schedules, airport capacity and operational constraints. | The public record does not disclose the model architecture or detailed performance thresholds. |
| Output | Predictive views and scheduling suggestions intended to reduce congestion and route around disruptions. | Suggestions still move through existing FAA and local facility review. |
| Initial scope | Limited use in Washington-area airspace, followed by a planned gradual expansion. | The FAA has not published live delay-reduction results from the rollout. |
The vendor is Air Space Intelligence. In June, the FAA announced that it had selected ASI for two linked systems: Flow Management Data and Services, known as FMDS, and SMART. The agency described FMDS as the data backbone for the Air Traffic Control System Command Center and SMART as a layer that uses that data to coordinate schedules and trajectories before aircraft depart.
Air Space Intelligence said the award is a 12-year, $875 million software and AI contract for FMDS and SMART. That June award is older than the freshness window, but it supports the current September rollout because it identifies the responsible vendor and scope.
Why limited mode matters
The careful rollout matters because air traffic management is a safety-critical setting. Ars Technica reported on September 18 that government and industry officials expected SMART to debut in the Washington area, and that airline concerns had eased after the FAA described the tool as producing alternative route information through existing processes rather than changing controller or airline procedures.
That distinction is central to the story. A system that predicts congestion can help managers plan departures, routings and recovery from weather disruptions. It can also create new accountability questions if a recommendation is wrong, data quality degrades or staff trust the display more than the conditions warrant.
Nextgov/FCW also reported that the D.C.-area airports are the first to receive the capability and cited the FAA fact-sheet language about predicting traffic flows and potential conflicts. The open evidence supports a measured claim: the FAA has moved SMART from planned deployment into limited operational use. It does not support a claim that AI is separating aircraft, replacing controllers or proving nationwide delay reductions.
What to watch next
The next useful evidence will be operational rather than promotional: how SMART performs during bad weather, how often facilities accept or reject its recommendations, whether delays fall on comparable days, and what criteria the FAA uses before expansion beyond Washington-area airspace.
For passengers, the promise is fewer cascading delays. For controllers and traffic managers, the practical question is whether a shared predictive view reduces workload without creating blind trust in a model. The FAA's first SMART deployment is small enough to test that boundary in public view.
Comments ()