Assaia, a Zurich company, turns an airport's existing CCTV into a live record of every aircraft turnaround, timing each task and warning operations teams when a stand falls behind. Its ApronAI system runs at London Heathrow (LHR), across 116 gates.
Founded in 2016 with 51 to 200 staff, it raised a 26.6 million US dollar Series B in December 2025, led by Armira Growth. ApronAI uses computer vision to log fuelling, baggage, boarding and pushback, then predicts off-block time; the same video core feeds tools for stand and resource planning, safety monitoring and engine emissions on stand.
Heathrow's rollout covers more than 540 cameras in Terminals 2, 3 and 5, and Copenhagen (CPH) monitors auxiliary power unit emissions with it. Dubai International (DXB) has contracted it for all stands, Munich (MUC) for 150 and New Chitose (CTS) for Japan's first full deployment. Breeze Airways signed for an initial 30 gates in August 2026.
Aviation AI deployments and updates
01 / Deployed
Assaia ApronAI turnaround analytics at Seattle-Tacoma with Alaska Airlines
Seattle-Tacoma uses artificial intelligence from Assaia to monitor aircraft turnarounds, and Alaska Airlines says its teams are already using the system, according to Assaia.
The vendor's numbers are large. In its Alaska customer story, Assaia claims a 17% on-time improvement, ground delays on delayed flights cut by 3.9 minutes on average, turnarounds 12% shorter, ground delay down 25% and more than USD 15 million in savings.
A separate Assaia story says SEA has cut taxi-in time by almost a minute per flight.
Hokkaido Airports will use Assaia's ApronAI at eight international stands at New Chitose, with operations planned from December 2026, the operator said on 30 June 2026. Assaia, announcing the deal the same day, called it the first ApronAI installation with full Japanese-language support, facilitated by Marubeni.
TRAICY reported on 26 July that the agreement had been signed. Regional Gateway covered it on 2 July.
Dubai Airports has chosen Assaia's AI turnaround-management system for Dubai International and says it "will be implemented to cover all aircraft stands", according to Assaia's announcement of 20 November 2025. The work is to be coordinated with Emirates, flydubai, dnata and other partners, with the stated aims of fewer delays and better on-time performance.
Assaia says the system will make Emirates the only airline in the world to use full-scale AI turnaround management across its entire operation at DXB. Majed Al Joker, Chief Operating Officer at Dubai Airports, and Nick Moore, Senior Vice President Passenger Services and Hub Operations at Emirates, are quoted. Airport Technology covered it on 24 November 2025 under a headline saying Dubai "deploys" the system, but the body of that report says it will be implemented.
No stand count in service, go-live date or turnaround results had been published as of 16 October 2026. flydubai's separate ZestIoT turnaround platform, also at DXB, is a different record.
Copenhagen Airport went live with Assaia's EmissionsControl on 9 September 2025, an AI-powered system that uses cameras to monitor auxiliary power unit use across the majority of its stands, according to Assaia. Assaia says this makes CPH the first airport in the world to do so, and Passenger Terminal Today reported on 15 September 2025 that CPH states it is the first airport in the world to achieve this.
What the airport gains is a round-the-clock, year-round view of APU use on stand.
The project supports Copenhagen's commitment to net-zero carbon emissions from its own operations by 2030, Assaia says.
Munich Airport has selected Assaia's ApronAI to monitor aircraft turnarounds across 150 stands initially, with further expansion planned, Assaia announced in August 2025. ApronAI applies video analytics, computer vision and machine learning to stand cameras to detect turnaround milestones and send alerts to operations teams, by Assaia's description.
Thomas Hoff Andersson, Munich Airport's chief operating officer, called it "a critical step towards enhancing our operations using data and artificial intelligence", as quoted by Airport Technology on 14 August 2025. Munich served nearly 42 million passengers in 2024, according to Assaia.
This is a selection, not a confirmed installation. Neither Munich Airport nor Assaia has said when the first stands will go live or published results.
Assaia's ApronAI watches aircraft turnarounds on British Airways stands at Heathrow Terminal 5. In the first phase, run with BA, 52 cameras were installed across 17 stands, roughly three per stand covering both the stand and the jetty, Assaia said in July 2025. Computer-vision models track tasks such as baggage unloading, refuelling and boarding from those feeds in real time, giving Heathrow and airline teams a running view of each turnaround.
The wider programme is still being installed. Heathrow, IAG and Assaia announced in July 2025 plans to extend ApronAI to 116 gates across Terminals 2, 3 and 5, with more than 540 cameras, by Assaia's figures. Heathrow said in December 2025 that its stand camera rollout would continue across all terminals with full coverage targeted for the end of 2026; that release does not name Assaia.
The relationship started in IAG's IAGi accelerator programme. Assaia says IAG has since made an equity investment in the company, and BA remains the airline partner at Terminal 5.
Neither Heathrow nor Assaia has published how many of the 116 gates were live as of 15 October 2026. There are also no published Heathrow turnaround results attributable to the system, such as changes in delay minutes or stand occupancy.
In April 2024 Fiji Airways said it had partnered with Assaia to implement TurnaroundControl, a computer-vision tool for monitoring aircraft turnarounds, BAE Ventures reported on 30 April 2024.
Berlin Brandenburg introduced Assaia's camera-based TurnaroundControl system in March 2024 to monitor aircraft handling and help ground teams coordinate turnaround activities. The first phase covered 16 stands, with another 33 planned. The airport confirmed the system was operational, but the additional rollout and measurable time savings require separate confirmation.
Rome Fiumicino uses Assaia's ApronAI to monitor aircraft turnarounds and predict off-block and ready times. Assaia's customer story says the system went live across 57 gates that handle about 75% of passenger traffic, with the remaining stands being rolled out.
ADR Ventures, the airport operator's investment arm, joined Assaia's Strategic Partner Community as an investor, Assaia said on 18 December 2023. Assaia credits the system with a six-minute cut in delays and EUR 18.8 million in value at Fiumicino; these figures are the vendor's own.
On 14 March 2023 the GTAA announced Assaia ApronAI at all Toronto Pearson gates. Assaia reported a 44% reduction in average taxi-in time at Pearson, in an undated customer story.
JFKIAT, which runs Terminal 4 at JFK, deployed Assaia's ApronAI TurnaroundControl there, Airports International reported on 1 December 2021. Computer vision tracks each turnaround, and the system flags excessive APU use.
Assaia makes bigger claims in an undated case study: a five-minute cut in delays, a 25% improvement in gates being clear for arriving aircraft and potential savings of USD 40 million a year.
British Aviation Group reported on 17 September 2019 that Gatwick was the first airport to deploy Assaia's apron AI computer vision for operational use.
ApronAI is Assaia's core AI platform that converts ramp camera video into structured operational data using computer vision and machine learning.
The problem it addresses
Airports, airlines, and ground handlers lack real-time visibility into apron operations, leading to undetected delays, poor on-time performance, inefficient gate utilization, and reactive rather than proactive decision-making. ApronAI replaces manual observation and fragmented data with automated, AI-generated turnaround event data.
Key features
Real-time turnaround event timestamping via computer vision
Predicted Off-Block Time (POBT) using live turnaround data, flight details, and weather inputs
Predicted Aircraft Ready Time (PRDT)
Live video feeds from aircraft stands, aerobridges, and apron
7 more features
Airline-specific business logic and alerting rules
Management-by-exception alerting to prioritize issues
Real-time and historical data dashboards
CSV data export
Multi-stand and multi-gate monitoring
Integration with Airport Collaborative Decision-Making (A-CDM) platforms
Open API for connection to AODB, dispatch tools, and business intelligence systems
Technical specifications (9)
Computer vision and deep learning algorithms processing CCTV/IP camera video streams
Machine learning for event prediction and anomaly detection
On-premise server deployment option (confirmed at CVG)
NVIDIA Jetson AGX Xavier modules for edge inference (confirmed at CVG deployment)
NVIDIA Metropolis intelligent video analytics platform (confirmed at CVG)
Compatible with existing camera infrastructure, no proprietary cameras required
Approximately 3 cameras per stand at Heathrow T5 phase 1 (52 cameras across 17 stands)
Scales to 150+ stands (Munich MUC initial rollout)
Scales to 540+ cameras across multiple terminals (Heathrow full rollout)
TurnaroundControl is a specialized interface application built on top of ApronAI that provides airlines, ground handlers, departure coordinators, and gate managers with full operational control and visibility over the aircraft turnaround process.
The problem it addresses
Airlines and ground handlers lack a centralized, real-time interface to monitor and manage multiple concurrent turnarounds, leading to missed alerts for process deviations (e.g., late catering or fueling), delayed push-back, and reactive delay management.
Key features
Multi-gate turnaround dashboard with color-coded status widgets
Watchlist to monitor selected turnarounds with live video
Airline-specific business logic and alerting (e.g., fueling, cleaning start/end)
Instant notifications for process deviations
8 more features
Centralized management of multiple simultaneous turnarounds
Predicted Off-Block Time (POBT) and Predicted Aircraft Ready Time (PRDT) display
Departure queue management based on POBT and PRDT
Integration with boarding progress and bag scanning data
Multi-airport map view
High-speed playback for post-event review and safety training
Support for A-CDM data contribution
SMS alerts for managers on required actions
Technical specifications (6)
Powered by ApronAI computer vision technology
AI and machine learning for turnaround milestone prediction
3.9 min average reduction in ground delays for delayed flights (Alaska Airlines deployment)
17% increase in on-time performance (Alaska Airlines deployment)
2 min average reduction in ground delay (United Airlines deployment)
Deployed across 106 gates at Toronto Pearson (full rollout)
SafetyControl uses the same computer vision infrastructure as ApronAI to detect unsafe situations on the ramp caused by SOP deviations, such as an aerobridge not fully parked.
Key features
Real-time detection of SOP deviations on the ramp
Colour-coded safety indicators across all active turnarounds
Mobile alerting for unsafe situation notifications
Airport map overview
2 more features
SSO support
Data export and API access
Technical specifications (1)
Powered by same computer vision stack as ApronAI Base. Requires existing CCTV camera infrastructure or ApronAI deployment.
StandManager
StandManager is an AI-powered Resource Management System (RMS) for airport stand and gate allocation.
The problem it addresses
Airports historically relied on fixed buffer times, scheduled flight data, and manual processes for stand and gate allocation, resulting in suboptimal utilization of constrained gate capacity. StandManager automates allocation using live data and intelligent, flight-specific buffers to increase effective stand capacity without new infrastructure.
Key features
Automated stand and gate allocation using live operational data
Analysis as a Service is Assaia's managed analytics engagement in which Assaia works directly with customers to build bespoke data assets, establish operational benchmarks, and generate custom KPIs from turnaround data captured by ApronAI.
Key features
Bespoke data asset creation from ApronAI turnaround data
Custom KPI definition and tracking
Operational benchmarking against peer airports or historical baselines
Collaborative engagement, Assaia works alongside customer teams
1 more feature
Outputs structured reporting for internal and external stakeholders
Technical specifications (1)
Delivered as a managed service on top of ApronAI data. Requires active ApronAI deployment. Output format tailored to customer requirements.
EmissionsControl monitors auxiliary power unit (APU) engine usage at aircraft stands in real time, providing data on APU runtime to support emissions compliance and carbon reduction programmes.
Key features
Real-time APU engine runtime monitoring per stand
Emissions compliance tracking and reporting
Supports carbon reduction and sustainability targets
Alert capability for excessive APU usage
1 more feature
Uses existing ApronAI camera infrastructure, no additional hardware
Technical specifications (1)
Runs on existing ApronAI camera and sensor infrastructure. No additional hardware required where ApronAI is deployed.
Customers and sites the vendor lists for this product. Aerostack has not verified these unless a deployment record is linked above.
Register notes
Notes held in the Aerostack register. Aerostack has not independently verified them.
Market Position
Turnaround delays remain significant contributor to flight delays industry-wide
Limited real-time visibility historically available to ops teams
Assaia leverages existing camera infrastructure, reducing deployment cost vs sensor-based alternatives
Competition
ADB SAFEGATE (SafeControl)
SITA (turnaround management)
Emerging CV startups
Differentiation
Installed base at major hubs
Airline partnerships
No additional hardware required Deployments include JFK, LHR, DXB, YYZ. Active Gatwick trial underway.
Earlier listing summary (3 groups)
Summary text from the earlier Aerostack listing. Figures in it are not independently verified.
Core Technology
Computer vision platform using ML-based object detection
Connects to existing CCTV infrastructure (Axis, Genetec, Milestone)
Automated timestamps for turnaround events
POBT prediction based on observed activity
Real-time alerting when events fall behind schedule
Deployment
35+ airports globally
Major hubs: JFK, Heathrow, Dubai, Seattle-Tacoma
Airline customers: United, Alaska, Southwest
Airport operators: Fraport, JFKIAT, Aeroporti di Roma
Business Impact
5-minute average reduction in ground delays reported
17% improvement in OTP reported by customers
One million monitored turnarounds reached November 2024
Profile data from the Aerostack companies register; deployment records from the verified register. Evidence grades: T1 first-party confirmation, T2 credible reporting. See Our approach for how records are graded.
Aerostack companies register
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.
