NTSB attributes preventable Washington midair collision to systemic FAA routing failures

NTSB Chairwoman Jennifer Homendy attributes the January 27, 2025 Washington midair collision—killing all 67 aboard an American Airlines jet and an Army Black Hawk helicopter—to systemic FAA routing failures rooted in a boundary structure that excluded risk-bearing parties from the decisions determining their exposure.
The NTSB locates the root cause not in the proximate 75-foot vertical separation or the contributing routing decision, but in the institutional boundary structure in which frontline controllers, affected passengers, and prior NTSB recommendations lacked standing to compel reconsideration.
NTSB investigator Katherine Wilson documented that controller workload reached saturation as traffic climbed from 10 to 12 aircraft in the final 90 seconds, while collision alerts calibrated to 900 feet provided inadequate reaction time at the policy-mandated separation margins.
The NTSB record shows that a 2013 Reagan near miss, a denied 2023 regional supervisor request for reduced traffic, and prior NTSB recommendations appeared as inputs that produced no commensurate regulatory action, exemplifying the FAA’s exclusion of the parties best positioned to detect rising risk.

The National Transportation Safety Board on Tuesday determined that the midair collision on January 27, 2025, near Washington, D.C., was “100% preventable.” All 67 people aboard both aircraft died, including 28 members of the figure skating community; it is the deadliest plane crash on U.S. soil since 2001. The collision involved an American Airlines jet from Wichita, Kansas, and an Army Black Hawk helicopter. NTSB Chairwoman Jennifer Homendy stated that the crash resulted from systemic failures rather than individual pilot error. The NTSB record, as reported, indicates that the parties bearing the risk of routing decisions at Reagan National Airport — the passengers aboard the American Airlines flight, the 28 figure skating community members among the dead, and the families who attended Tuesday’s hearing — had no standing in the decisions that determined their exposure.

Boundary Structure and Risk Allocation

Under the boundary-critique framework developed by systems theorist Werner Ulrich, the NTSB record surfaces four categories of affected-but-not-involved parties across the system’s structure. Motivation: The documented sequence of decisions — maintaining 36 hourly arrivals at Reagan until the crash — indicates a systemic prioritization of throughput, while the parties bearing the risk of those throughput decisions had no standing in the routing decisions that determined their exposure. Control: The regional supervisor who requested reduced traffic in 2023 was, according to the NTSB record, denied; decision-making authority over route configuration remained at a level where frontline controllers’ experiential knowledge did not translate into binding input. Knowledge: Controllers’ real-time awareness of workload saturation — NTSB investigator Katherine Wilson characterized the controller’s “reduced situational awareness” — became legible to investigators only after 67 deaths, rather than serving as the kind of operational signal that triggered route adjustment. Legitimacy: The 2013 near miss at Reagan, the 2023 regional supervisor request, and the NTSB’s own prior recommendations on similar hazards all appear in the record as inputs that did not, in the NTSB’s characterization, produce commensurate regulatory action. Systems theorist Werner Ulrich’s framework characterizes the FAA’s failure to act on the 2023 request as the predictable output of a boundary structure in which the parties best positioned to detect rising risk are excluded from the decisions that would address it.

The findings identify three levels of causation. Proximate cause: the 75-foot vertical separation in shared airspace, produced by a helicopter route configured against aircraft landing on Reagan’s secondary runway. Contributing cause: the specific routing decision that produced that separation. Root cause: the institutional boundary structure in which such routing decisions are made without the parties who bear their consequences having standing to compel reconsideration. Removing the routing decision would have prevented the collision at this location. Revising the institutional structure that produces such routing decisions — what Ulrich’s framework names the boundary judgment between “involved” and “affected-but-not-involved” — would be required to prevent recurrence at other airports facing similar geometry.

Causal Architecture and Cognitive Limits

Two concurrent causal chains converged in the collision. Policy chain: Institutional policy (maintaining 36 hourly arrivals; routing helicopters with 75-foot vertical separation) resulted in environmental conditions in which 12 aircraft occupied a single sector. Technological chain: Collision alerts calibrated to 900 feet resulted in inadequate reaction time at the reduced separation margins. NTSB research indicates that alerting pilots at 300 feet rather than the current 900 feet could eliminate 90 percent of near misses; a bill endorsed by Homendy would require aircraft to carry advanced locator systems — a measure the NTSB has recommended for years. The collision occurred at the intersection of these chains: policy-mandated density overwhelmed the controller’s capacity, and the technological safeguards failed to provide adequate reaction time.

Wilson reported that an air traffic controller felt overwhelmed when traffic reached 10 aircraft about 10 to 15 minutes before the collision. Traffic volume then increased to 12 aircraft — seven airplanes and five helicopters — about 90 seconds before impact, at which point the controller’s workload, in Wilson’s characterization, “reduced his situational awareness.” The system, per the NTSB record, did not include a mechanism for automatically reducing sector density when controller workload reached capacity; the operational posture reflected a dynamic in which short-term scheduling continuity was maintained by shifting the burden of complexity onto the individual controller rather than redesigning the structural environment to cap volume at safe thresholds.

A reinforcing loop dominated the period from 2013 to January 2025: throughput pressure increased controller workload, which degraded situational awareness and produced near misses, which generated warnings, which the FAA received without translating into route changes, which preserved throughput capacity and therefore the pressure on the next controller. The balancing loop — NTSB recommendations generating public attention and FAA response — operated on a delay long enough that the public attention required to close it accumulated primarily through fatal incidents. Systems theorist Peter Senge identifies this pattern as a “limits to growth” archetype, in which a reinforcing growth process overshoots its balancing constraint when the balancing response is delayed.

Attributed Characterizations of Institutional Conduct

The hearing record contains specific attributed characterizations of the institutional conduct. NTSB Chairwoman Jennifer Homendy stated, “We should be angry. This was 100% preventable. We’ve issued recommendations in the past that were applicable to use. We have talked about seeing and avoid for well over five decades. It’s shameful.” Former Transportation Department Inspector General Mary Schiavo characterized the agency’s inaction, stating, “It was just a shocking dereliction of duty by the FAA. And they have so much work to be done to fix it.” Kristen Miller-Zahn, whose brother was killed, addressed the tragedy as institutional negligence: “The negligence of not fixing things that needed to be fixed killed my brother and 66 other people.”

Regulatory Response and Unresolved Variables

In response to the findings, the NTSB issued more than 30 recommendations on Tuesday, the majority directed at the FAA, covering controller training on visual separation, staffing levels, and a comprehensive reevaluation of helicopter routes at Reagan and other airports. FAA post-crash measures include: hourly plane arrivals at Reagan reduced from 36 to 30; tower staffing increased to 22 certified controllers with 8 more in training; permanent changes made to keep helicopters and planes out of the same airspace around the airport. The bill endorsed by Homendy (advanced locator systems) and the 300-foot alert threshold (which NTSB research ties to a 90 percent near-miss reduction) address controller-side and equipment-side variables.

These measures leave unresolved the institutional mechanism by which a future regional supervisor’s traffic-reduction request, or a future controller’s reported overload, would be translated into binding route revision before a fatal outcome rather than after. The NTSB’s recommendations operate within the existing boundary structure. Whether the structural question of who counts as “involved” in routing decisions is itself reopened is a question the hearing record, on its face, does not resolve.

NTSB statistics show 1,405 crashes nationwide — the lowest number since the pandemic in 2020 — despite several high-profile crashes and near-misses in the months since the collision.

Additional Considerations

Regarding surfaced tensions and gaps in the investigative record, the substrate focuses primarily on Federal Aviation Administration institutional failures and routing decisions. The specific procedural actions of the Army Black Hawk crew in the moments before the collision are not detailed in the available hearing record; treatment of the Black Hawk crew’s actions is a noted gap, not a deliberate scope-clip. Additionally, regarding the analytic scope of motive theories, whether the constraint against motive theories extends in this mode to inferences about an institution’s operating priorities, or applies only to named natural persons, remains an interpretive question. The corpus reframes the relevant claim as documented conduct (the 36 hourly arrivals maintained until the crash), removing the question from the load-bearing path; the underlying interpretive question persists for downstream consumers.

Analytical techniques used in this piece

This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.

Boundary Critique: Examines whose interests and voices the framing of a problem includes — and whom it leaves out.
Root-Cause Analysis: Traces a symptom back along its causal chain to the conditions that actually generated it.
Systems Dynamics (Causal): Models the feedback loops and delays that drive a behavior over time.