How AI Optimizes Gate Allocation

The Obstacle: Delays and Lost Revenue

To operate smoothly, airports rely on efficient allocation of stands and gates to aircraft. Avinor, a state-owned entity that manages 43 of Norway’s airports, recognized an opportunity to improve this critical process. The problem: at Oslo Airport, apron coordinators responsible for gate allocation - while seasoned professionals with deep expertise - relied on legacy IT systems and manual processes to analyze the complex, multivariate optimization problem of allocating gates in such a massive, bustling airport. The combination of legacy systems and manual processes presented opportunities to further enhance the efficiency and quality of decision-making in gate allocation, especially when managing real-time changes.

Our Approach: AI to Optimize Decisions

Avinor and BCG set out to apply AI-driven solutions to the gate allocation challenge, developing a model that enhances apron coordinators’ ability to make more informed and faster decisions while focusing on key operational priorities. The most important value driver was to increase capacity utilization by finding the space for another flight during peak hours. Besides that, better gate allocation would reduce costs by requiring fewer busses to transport passengers between terminals and remote stands. In addition, the model could improve commercial revenue by analyzing passenger profiles and placing high-value travelers near gates with quality retail outlets. For the long term, the model could be applied to strategic requirements, such as testing new flight schedules and delaying terminal expansions and other costly CAPEX investments.

Working together over a ten-week period, BCG and Avinor’s new AI team deployed a Proof of Concept (PoC) that captures the complex logic and expertise apron coordinators rely on when making gate allocation decisions. To develop the model, the Avinor team incorporated the most important of 20,000 constraint hard and soft rules that the solution must satisfy, using machine learning to predict potential delays and Monte Carlo simulations to assess gate overlap risks.

Using data visualization techniques, the team developed a tool that generates gate allocation optimization plans with insights into both delay and early-arrival probabilities, flight overlap risk, and passenger profile information. And while providing recommendations, the tool also allows human experts to make adjustments.

In addition to enhancing day-to-day operations, “what-if” simulations support long-term strategic planning such as the feasibility of adding new flights to existing schedules.

The Results: Cost Savings and New Insights

The PoC clearly demonstrated substantial value in optimizing stand and gate allocation operations, with the analysis indicating significant annual value potential across several areas, including the following:

• Improved capacity utilization, by enabling the scheduling of additional peak-time flights.
• Reduced costs from fewer disturbances and delays, driven by a lower risk of gate overlaps.
• Increased non-aeronautical revenue, through improved alignment of gate allocations that include passenger interests and are in close proximity to commercial areas.
• Lower ground transport expenses, achieved by reducing the need for bus transfers between aircraft and terminals.

Overall, the findings confirmed that data-driven gate allocation can unlock meaningful operational efficiencies and financial benefits while enhancing the passenger experience.

Avinor’s AI team also gained valuable hands-on experience and assumed full ownership of the model - demonstrating the power of AI to augment human experience and knowledge.

Beyond its immediate benefits, the PoC represents a significant step toward Avinor’s strategic objective of achieving AI-driven airport operations under a Total Airport Management model. The initiative also proves that AI can balance cost reduction with revenue enhancement, and be applied to specific tactical problems as well as long-term strategic goals. Through a collaborative effort, the BCG/Avinor partnership has delivered a scalable, strategic tool that empowers employees and leadership to make better decisions, faster.