As aviation nears capacity limits, reliable prediction depends on trusted, shared data. Geneva Airport combines its AOP (airport operations plan), AI-based forecasting and DCB (demand capacity balancer) to anticipate constraints, optimise resources and support co-ordinated, proactive decision-making within the A-CDM (airport collaborative decision making) framework.
From reaction to anticipation
European aviation is entering a period in which operational performance will increasingly depend on the ability to see constraints before they materialise. In 2025, the network handled around 11 million flights, yet only 76% achieved 15-minute on-time performance. Nearly half of all delays were reactionary, showing how a local disruption can quickly propagate across rotations, resources and the wider network. In this context, the Airport Operations Plan (AOP) is not just another planning layer; it is a way to build a shared forward-looking view so airports can anticipate pressure points instead of simply responding to them.
At Geneva Airport (GVA), this represents an evolution from fragmented information towards increasingly co-ordinated decision-making. Before A-CDM, information was not consistently updated between operational stakeholders, and co-ordination between flight, resource and ground activities was limited. A-CDM introduced a shared short-term reference. The AOP will extend this approach towards a rolling six-month operational view, shared with the local community and the Network Manager.
Building a common operational truth
Before an airport can predict future constraints, it must first establish a shared understanding of current operations. A well-structured AODB (airport operational database), supported by reliable data sharing between airport partners, provides this common reference.
This common operational truth relies on three principles: information must be accessible, source data verified and validated, and calculations fully traceable. These principles build trust in the data and ensure that forecasts, indicators and decisions are based on the same operational reality.
Use of AI for forecasting
Traditional forecasting is based on flight plans available for the current season. Combined with seat-load-factor forecasts and booking information provided by airlines, this method delivers high-quality short-term results over a period of a few weeks.
However, plans for the next season are available only a few weeks before the season starts. This impairs mid-term forecasting up to 12 months ahead. Combining this information with past data and expected developments is highly challenging when done manually.
To overcome this challenge, we developed an AI-based mid-term forecasting tool that outperforms traditional forecasting methods.
Translating forecasts into action
Forecasts are valuable only if operational teams can make direct use of them. A Demand Capacity Balancer (DCB) transforms forecast information into actionable operational insight.
Given that most commercial products are “black-box” solutions, we developed an internal tool. This allows us to retain control over the calculations and algorithms.
The DCB compares future demand with future capacity and converts mathematical outputs into simple indicators. A severity score reflects passenger experience through waiting-time thresholds, while a stress index measures operational margin by comparing demand with capacity and identifying periods approaching or exceeding 90% utilisation. The aim is to make predictive information understandable at a glance.
This approach enables sharing a comprehensive rolling six-month plan with all stakeholders.
Conclusion
The transition from reaction to anticipation is now a strategic necessity for airport operations. It cannot rely on isolated tools or individual expertise alone. It requires a strong and trusted data foundation: reliable information, shared definitions, transparent calculations and effective collaboration between all operational stakeholders.
When these elements are combined with advanced forecasting and demand-capacity analysis, data becomes more than a reporting asset. It becomes an operational decision-support capability, helping airports identify future constraints earlier, test possible responses, allocate resources more effectively and align local decisions with the wider network.
Ultimately, the value of prediction is not to forecast the future for its own sake, but to create the visibility, confidence and shared understanding needed to act early. For Geneva Airport, the AOP, AI-based forecasting and the DCB form a coherent step towards more resilient, co-ordinated and proactive airport operations.
I will be speaking on this very topic during my session ‘From firefighting to foresight: building the data foundation for prediction’ at International Airport Summit 2026 taking place in Rome on 10-12 November. Make sure you join my session to discuss one of the industry’s most important challenges and network with me afterwards.
Register for your FREE* ticket to attend the summit.
* Free aviation leader tickets are applicable to senior managers and above from airports, airlines, regulatory bodies and aviation authorities.
No comments yet