NATS has traced the air traffic control failure that disrupted hundreds of thousands of passengers this month to a software defect in a small section of the system used to manage UK airspace.
A preliminary investigation into the incident on 8 September found that corrupted flight data was generated after two processes interacted in an unexpected sequence. The failure occurred within a millisecond, but the resulting reduction in information available to controllers led to restrictions on air traffic across the UK for around six hours.
NATS said the defect sat within the National Airspace System, which underpins the management of UK airspace. The affected software is used when aircraft identification codes need to be allocated manually rather than through the normal automated process.
During the incident, the system was processing one of those manual requests when it received a higher-priority message. Processing of the aircraft-code request was temporarily paused, but a defect meant it failed to resume correctly when the higher-priority task had been completed.
The resulting output was corrupted and subsequently affected some further flight-data updates. Controllers therefore had less information available than normal, prompting NATS to restrict the number of aircraft entering the system while the technology was restarted and flight information reloaded.
The immediate software problem affected flights handled by the London Area Control centre, primarily higher-level traffic operating above 24,500 feet. Restrictions nevertheless had to be imposed more widely to allow the national system to be recovered safely.
NATS chief executive Martin Rolfe said: “This was a software issue in a specific part of our flight data system, that we have traced to a small subsection of coding.”
The company has put mitigation measures in place while a permanent fix is safety-tested before deployment. NATS also said the September incident was unrelated to the major system outage experienced in August 2023 and rejected speculation that military activity had caused the latest failure.
Operations returned to normal on the evening of 8 September, but disruption continued for more than two days as airlines, airports, and passengers worked through the backlog created by cancelled and delayed flights.
The preliminary findings materially advance the explanation available when Business Quarter reported the initial disruption. At that stage, the operational consequences were established but the specific technical mechanism had not been identified.
The incident renews attention on the resilience of critical infrastructure built on complex software estates. In air traffic control, the consequences of an isolated defect can extend rapidly through airlines, airports, ground handling, passenger connections, freight movements, and crew scheduling even when the core technical failure lasts only a matter of hours.
NATS emphasised that safety was not compromised. Restricting traffic is itself part of the system’s protective response when controllers do not have the level of information required for normal operating capacity.
The commercial consequences of such restrictions can still be substantial. Airlines face additional accommodation, rebooking, crew, fuel, and positioning costs, while airports can experience congestion as aircraft and passengers are displaced from planned schedules. Disruption can persist after the technical service has recovered because aircraft and crews are no longer in the locations required for subsequent flights.
The findings also put software assurance at the centre of the response. The defect depended on a specific sequence involving a manual code request and a higher-priority activity, illustrating the difficulty of testing systems against rare interactions that may not occur during normal operation.
NATS said it has invested more than £1bn in systems and technology over the past decade and is preparing plans for the Civil Aviation Authority covering a further £1bn of investment by the end of 2033.
The full investigation is still to come. The document published on 18 September is explicitly preliminary and was produced after a request from Transport Secretary Heidi Alexander. Further work will examine the incident in more detail, while the permanent software correction must undergo safety testing before it can be introduced.
The September failure is separate from the August 2023 outage, which arose from a different technical problem. Two substantial disruptions within three years nevertheless increase scrutiny of resilience across systems where even short periods of reduced capability can propagate through an entire transport network.
NATS has established the immediate technical cause, but its final investigation and the implementation of the permanent fix will determine how the incident changes future software assurance and investment priorities.
You must be logged in to post a comment.