Every major airport operates as a complex network of aircraft, vehicles, personnel, infrastructure, weather systems, and emergency services. Each produces information that could help prevent accidents. Yet much of that information remains separated across departments, databases, and operational systems.
What if an airport could recognize a developing accident sequence before any individual department realized something was wrong?
That question drives The Intelligent Airport: Artificial Intelligence as the Next Layer of Airport Safety Management, a 16-page research investigation published September 3, 2026, by aviation safety researcher Albert N. Clark in the ASX Research Journal and Database.
The study proposes an integrated artificial intelligence layer capable of continuously evaluating airport-wide safety conditions. Rather than replacing controllers, airport managers, firefighters, or maintenance professionals, the technology would connect information that existing systems often examine independently.
Airports Have the Data. They Lack the Connections.
Modern airports already employ surface surveillance, weather monitoring, maintenance databases, vehicle tracking, runway inspection systems, wildlife reporting, and formal Safety Management Systems (SMS).
Individually, these technologies provide valuable information. Their limitation is fragmentation.
A deteriorating pavement section may not independently justify restricting operations. Neither might an approaching thunderstorm, a closed taxiway, reduced emergency-response availability, or an unusually heavy arrival schedule.
But when those conditions occur together, the resulting operational risk may be considerably greater than the sum of the individual hazards.
An intelligent airport would continuously examine those relationships, identifying combinations that could otherwise remain hidden between organizational boundaries.
Predicting Runway Conflicts Before They Develop
Runway incursions and surface collisions represent particularly promising applications for predictive artificial intelligence.
Traditional surveillance technologies, including surface radar, ADS-B, multilateration, and runway-status systems, provide information about aircraft and vehicle movements or warn of detected conflicts.
Machine-learning systems could add a predictive dimension.
The research examines emerging models that combine airport taxiway geometry, historical traffic patterns, and changing operational conditions to forecast potential conflict locations.
Instead of asking only where aircraft and vehicles are positioned, an intelligent system could evaluate where dangerous interactions are most likely to develop during the next several minutes.
Such predictions could help human operators intervene earlier, before a conflict reaches the point of requiring an immediate response.
Artificial Intelligence Could Detect Hazards Humans Miss
The potential extends well beyond aircraft movements.
Computer vision and machine learning are already being investigated for detecting foreign object debris (FOD), identifying pavement deterioration, and supporting airfield inspections.
The paper describes FAA proof-of-concept testing of AI-assisted FOD detection at Cape May County Airport and Atlantic City International Airport. The testing reported a 96% detection rate for specified debris items, although false positives, processing time, weather, and low-light performance remained important limitations.
Future integrated systems could also combine wildlife activity, standing water, lighting failures, drainage conditions, pavement wear, and construction restrictions into continuously updated hazard assessments.
Rather than distributing inspection resources uniformly, airport personnel could prioritize locations where several warning signs indicate elevated risk.
Weather Changes More Than Flight Schedules
Airport weather management is another area where integrated AI could improve situational awareness.
A thunderstorm can affect runway configuration, visibility, surface friction, ramp staffing, aircraft servicing, emergency access, gate availability, and ground-vehicle movement simultaneously.
Conventional systems often distribute those consequences among different operational teams.
An intelligent airport could model the combined effects, allowing decision-makers to see how a weather event changes the safety condition of the entire airport rather than simply its arrival and departure capacity.
Protecting the People Who Keep Airports Running
The research also examines how staffing, fatigue, workload, and emergency-response readiness contribute to airport safety.
AI could potentially recognize when a heavy arrival period coincides with limited personnel, reduced rescue-and-firefighting resources, deteriorating visibility, and restricted taxiway access.
However, integrating human-performance information introduces serious ethical questions.
Fatigue monitoring, biometric data, employee surveillance, and predictive performance assessments could threaten privacy and workplace rights if implemented without appropriate safeguards.
The study therefore argues that intelligent systems should evaluate operational risk without turning airport employees into permanently monitored subjects of automated judgment.
Industry Is Already Building the Components
Several major aviation technology companies are developing capabilities related to the intelligent-airport concept.
The research identifies ADB SAFEGATE’s CORTEX platform as a particularly relevant commercial precursor, integrating airport operational functions through a common software architecture.
Saab, Honeywell, and SITA are also advancing technologies involving surface surveillance, airside operations, situational awareness, and shared operational information.
According to a SITA industry survey cited in the paper, 73% of airports were investing in artificial intelligence in 2025.
Nevertheless, the research distinguishes these developing capabilities from a fully integrated, airport-wide AI safety intelligence system. The individual technologies exist in varying stages of maturity, but the complete architecture remains an emerging concept rather than an established operational standard.
Who Is Responsible When AI Gets It Wrong?
Integrating artificial intelligence into airport Safety Management Systems raises questions that extend beyond engineering.
If an AI system recommends closing a runway, changing an operational procedure, or escalating a hazard warning, airport officials must understand the evidence behind that recommendation.
They must also retain the authority to evaluate and act upon it.
The Federal Aviation Administration requires certain certificated airports to implement Safety Management Systems, but it does not mandate or approve a particular SMS software product.
Airport operators remain responsible for their safety obligations, regardless of the technology supporting their decisions.
Consequently, future systems will require transparent risk calculations, reliable records, cybersecurity protections, documented human oversight, and clear accountability when recommendations are accepted, rejected, or ignored.
The Future Airport Must Understand Itself
The study ultimately envisions an airport maintaining a continuously updated digital representation of its own operations.
Aircraft movements, vehicle locations, runway conditions, weather, maintenance records, construction activity, wildlife reports, staffing levels, and emergency resources could all contribute to a common safety model.
Artificial intelligence would search for emerging combinations of risk, while human professionals would remain responsible for operational decisions.
The objective is not an airport controlled by machines. It is an airport capable of recognizing dangerous relationships that its individual departments may never detect on their own.
The next major advance in airport safety may therefore come not from collecting more information, but from finally connecting the information aviation already possesses.
Read the complete research: The Intelligent Airport: Artificial Intelligence as the Next Layer of Airport Safety Management, Albert N. Clark, ASX Research Journal and Database, September 3, 2026. The original 16-page publication includes the full analysis and supporting scientific references.