Superlinear, a Brussels AI consultancy, builds forecasting models for Brussels Airport (BRU), including the prediction of when passengers' bags will reach the belt.
Founded in 2018 with 51 to 200 staff, it prototyped the luggage-retrieval model in four months for the airport's bTag service and reports 95 per cent accuracy. It then won the tender for the airport's AOP 2.0 programme, building forecasts for check-in, screening, assistance, baggage and border control from 2022. Brussels is its only named aviation client.
Aviation AI deployments and updates
01 / Deployed · limited scope
Superlinear machine-learning modules for Brussels Airport AOP 2.0
Superlinear, a Belgian data-science firm, says it won the public tender for data science on Brussels Airport's AOP 2.0 and has been building machine-learning forecasts for five use cases: check-in, security screening, passengers with reduced mobility, baggage and border control. The aim is to tell operations staff where and when bottlenecks will form so they can act in advance.
Superlinear reports a mean absolute percentage error of about 30% per 30-minute interval for its check-in forecast, 14% at day level for the number of passengers needing assistance, and about 28% per flight for bag counts. These are the vendor's figures, and the baggage work was described as still ongoing.
The case study says work on the use cases began in May 2022 and that the models "will soon be put into production". Superlinear now marks the page as reflecting earlier work, and Brussels Airport has not confirmed which of the five modules are live.
bTag tells passengers arriving at Brussels Airport when their bags will reach the reclaim belt. Superlinear built the prediction model with the airport's internal team in a project that began in June 2018, and the model takes account of where the flight has come from, the time of day and other conditions.
The prediction is sent to the passenger automatically with the walking time from the aircraft and the belt number. The model runs on serverless cloud infrastructure, using computing resources only when needed, Superlinear says.
Superlinear reports that 95% of predictions are accurate, that 90% of passengers rated the tool "excellent" or "very good" and that 35% changed what they did while waiting. All three are vendor figures, from a case study that now carries a note saying it reflects earlier work, and Brussels Airport has not published a current statement on bTag.
Airport Operations Plan (AOP) AI Forecasting Suite
An AI-powered forecasting suite built for Brussels Airport Company under the Airport Operations Plan (AOP), an EU co-funded initiative.
The problem it addresses
Airport operators and their service partners struggle to right-size staffing and resources across multiple passenger-facing processes, leading to queue overruns, understaffing, or excess cost. The AOP forecasting suite provides advance demand predictions to enable proactive, cost-efficient resource allocation.
Key features
Passenger arrival volume forecasting at check-in in 30-minute intervals
PRM passenger count forecasting per flight at day level (14% MAPE reported)
Baggage volume and transfer bag forecasting per flight (28% MAPE per flight reported)
Border control passenger flow forecasting across three of five airport border control locations
A machine learning model that predicts, at the moment of aircraft landing, the exact time a passenger's luggage will arrive on the reclaim belt, together with walking time from gate to reclaim area and the specific belt assignment. The prediction is automatically communicated to passengers.
The problem it addresses
Passengers arriving at an airport do not know when their luggage will be available for collection, causing stress and inefficient use of time in the arrivals area.
Key features
Real-time luggage arrival time prediction triggered automatically on flight landing
Accounts for flight origin, time of day, number of passengers, weather conditions, and luggage carrier specifics
Delivers belt number, walking time, and predicted collection time to passengers
Serverless cloud architecture for scalability and cost efficiency
3 more features
Integrated into Brussels Airport bTag product
95% prediction accuracy reported
90% of passengers rated the tool excellent or very good
Technical specifications (3)
Serverless architecture using cloud compute on demand
ML model triggered by automatic database event on flight arrival
Outputs written to passenger communication channel via bTag integration
Verified deployments
Brussels Airport (BRU) – bTag machine-learning bag-reclaim time prediction at Brussels Airport
An R&D-stage AI solution using a Constraint Programming Satisfiability (CP-SAT) model to optimize the allocation of check-in desks at Brussels Airport. The model takes flight schedules and passenger arrival patterns as input and produces desk assignments that minimize passenger queue overflow.
The problem it addresses
Suboptimal allocation of check-in desks leads to uneven queue lengths, passenger congestion, and underutilized resources at airport check-in.
Key features
CP-SAT model with over 831,000 variables and 1.73 million constraints
Minimizes passenger overflow outside queuing zones
Enables scenario simulation such as relocating check-in areas or adjusting desk opening times
Takes into account airline-specific requirements and peak time management
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.
