Istanbul: from crowd alerts to the operations centre
At Istanbul Airport, AI is reported in passenger-density monitoring, a shared operating dashboard and an assistant that answers traveller questions. The challenge is separating working functions from broader claims about efficiency and experience.
Aerial view of Istanbul Airport main terminal, December 2019. Photograph: Kulttuurinavigaattori / CC BY-SA 4.0 (cropped to fit). Shows the airport, not an AI implementation.
The picture at Istanbul Airport
The pressure of running a very large terminal
An airport can count people passing a security point and still be late to act on a crowd forming elsewhere. Istanbul has described camera-based density alerts at passenger processing points, giving operational teams a chance to respond to local congestion.
The airport's F1RST digital-twin and Smart Dashboard work takes a broader view, combining information from multiple systems for staff and equipment decisions. Passenger communications are another strand, with Sestek's virtual assistant answering questions through several channels.
These accounts come principally from airport executives, technology suppliers and trade reporting. The tools are described in use, but independent evidence of the airport-wide performance gains is much more limited.
Selected implementations
Three ways of seeing what passengers need
01 / Deployed
The crowd forms before the queue is reported
The operational problem
Passenger numbers at an entrance, check-in desk or passport control can become a problem before conventional queue measures show it.
What the sources report
IST Systems' general manager told AeroTime in January 2025 that Istanbul uses AI for passenger-density monitoring at entrances, security, check-in and passport control, with alarms sent to the relevant airport teams. International Airport Review later described AI-powered CCTV and crowd analytics.
Why it matters
The operational value depends on whether the alert arrives early enough for teams to adjust a lane, staffing or passenger route.
What remains unproven
Public sources reviewed do not establish camera coverage, alert precision, false-alarm rates or the effect on waiting time.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
02 / Deployed
What the digital twin shows the operations centre
The operational problem
Airport teams need a shared interpretation of events when staff, equipment and passengers compete for capacity.
What the sources report
Istanbul's IST Systems unit describes F1RST as a Smart Dashboard and digital twin integrating operational systems. Its general manager told AeroTime that AI supports real-time, multidimensional analysis for allocating people and equipment.
Why it matters
A dashboard becomes operationally useful when its output informs resource allocation. Claims of environmental or reporting savings should be tested against clear baselines and independent evidence.
What remains unproven
The AI components have not received an independent technical assessment; figures cited by staff in later academic interviews are not audited AI-specific benefits.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
03 / Deployed
Answering a passenger without another queue
The operational problem
A traveller may need immediate guidance without finding a staffed desk or waiting for a call centre response.
What the sources report
Sestek describes a virtual assistant supporting Istanbul Airport's app, WhatsApp and phone services, with other reporting identifying a wider set of channels. The supplier says it has answered more than a million visitor questions and saved agent hours.
Why it matters
The assistant's operational test is not the size of the question count but whether it supplies the right answer, resolves the traveller's problem and hands off when it cannot.
What remains unproven
Containment rates and independently measured service quality have not been published; performance figures are attributed to Sestek and the airport.
Other tracked work includes runway foreign-object-debris detection and mobile-app personalisation. The runway system has supplier evidence of installation but no published detection rates; the personalised offers have supplier-reported commercial gains without a clear measurement period.
SAS Customer Intelligence 360 personalisation in the iGA Istanbul Airport app · Deployed
iGA Istanbul Airport uses SAS Customer Intelligence 360, with SAS partner Partner Republic, to personalise offers and information in its mobile app, according to a SAS press release dated June 2024. SAS says the app now adjusts journeys by passengers' stress levels, guides shoppers with augmented reality and uses beacons to send real-time notifications at fast-track entry points.
SAS reports a significant increase in revenue and double-digit growth in participation in iGA Pass services. Its customer story puts the revenue increase at 15%.
Both figures are SAS's own and neither page states a measurement period. The consent and data-retention terms for this personalisation had not been published as of 16 October 2026.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
ArgosAI A-FOD foreign-object-debris detection, runway 18-36 at Istanbul Airport · Deployed
ArgosAI, a Turkish company, has sensor stations for its A-FOD foreign-object-debris detection system installed on runway 18-36 at Istanbul Airport. The company says it showed them to members of ACI's technical and safety committees during an airside visit linked to the 43rd ACI TOSC meeting, in a 2022 item on its announcements page.
ArgosAI describes A-FOD as its own image-recognition system, built with deep neural networks, for spotting debris on the runway surface in any weather. The system watches the runway from the sensor stations and is meant to alert the airport to objects that could damage aircraft.
The sources are ArgosAI's own pages. Detection rates, the number of stations and whether the system now covers other runways at Istanbul had not been published as of 16 October 2026.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
An alert is not the same as an outcome
Passenger density and airport resource allocation require action at a particular moment. The virtual assistant has a different test: whether a passenger gets useful guidance without having to start again elsewhere.
Istanbul's public records identify a broad set of systems. The next substantial evidence would show which decisions changed and what happened as a result.
Live data from the Aerostack register. Status reflects published evidence only. See Our approach for how records are graded.
Aerostack register · 5 tracked deployments
Aerostack tracks AI deployments across airports and airlines, with the published evidence behind each record. Photography is real aviation imagery. Record thumbnails may show the named airport or airline; other photographs do not.
