In this series of articles, our Product Manager, Justin Keene, explores the transition from traditional Resource Management Systems (RMS) to Total Airport Management.
In this 1 st instalment, he delves into using AI for precise off-block time predictions and operational optimisation.
Optimising the use of airport resources isn’t just about efficiency—it’s a strategic move to unlock all available capacity, accommodate growth and delay costly expansion. Yet the old way of planning, based on fixed schedules and airline preferences, falls short when faced with peak capacity constraints and the inevitable day of operations events. Every decision to reallocate gates has a ripple effect, impacting everything from on-time performance to emissions.
To overcome these challenges, we need to rethink gate management. Instead of rigid plans, we must go beyond the traditional approach and adopt agile strategies that learn from the past, adapt to real-time situations, and proactively and efficiently improve operations.
Navigating the day of operations challenges
Efficient resource and gate management hinge on precise turnaround times, with the Target Off Block Time (TOBT) at its core. Initially, based on flight arrival data, the TOBT is updated by airlines and ground handlers 30 to 40 minutes before departure. It’s then crucial that the TOBT estimate is available at least 10 minutes before departure to be relevant and useful.
The pressure to provide an accurate estimate is immense. It hinges on ground handlers’ experience with estimating – a challenge if they’re new to the job – and whether they truly have good visibility of the situation and all its moving parts.
As a result, many large airports report accuracy rates of less than 60% within a 5-minute window, with even fewer TOBTs hitting the mark within the critical 10 minutes. This lack of precision leads to apron congestion, gate unavailability, departure sequencing issues and broader air traffic flow impacts.
Embracing machine learning: Making precise pushback predictions
Acknowledging the constraints of traditional approaches, progressive airports are turning to machine learning and AI to fine-tune TOBT precision. This technology delves into past off-block time performances, considering factors like origin, destination, carrier, aircraft type, gate, passenger load, and airfield condition to enhance accuracy.
Projected off-block times aren’t static milestones. As a machine learning engine processes real-time data—such as live flight information, weather updates, boarding progress, passenger flow metrics, or airside video analytics—it dynamically adjusts predictions in response to the unfolding situation. This predictive ability helps airports anticipate delays, collaborate effectively, and adjust departure sequences as needed, continuously improving predictions based on past experiences.
Unlocking resource optimisation: Orchestrating efficiency and sustainability
Predicting off-block times isn’t just about turnaround accuracy; it’s also useful in helping planners allocate resources to avoid gate clashes and optimise operations. When this predictive data is integrated into resource planning systems, gate managers get early warnings about potential delays. This foresight allows them to act proactively, minimising disruptions.
Yet, the advantages of AI extend beyond mere alerts; it’s about smart analysis and recommendations. For example, if the machine learning model predicts a late off-block time, it alerts the planner about potential gate conflicts. The planner has options like holding the flight at its original gate or rearranging schedules to different gates.
Imagine if the system could further suggest the best reallocation, helping the planner assess the impact on passenger connections, overall apron efficiency, and carbon emissions. By offering augmented decision support, AI empowers planners to make informed decisions aligned with airport priorities and drive optimal performance.
Going beyond RMS toward Total Airport Management
Picture an airport where every piece falls into place seamlessly. That’s the promise of integrating machine learning into resource management—and it is a fast track towards Total Airport Management.
This strategy goes beyond resource allocation, aiming for the perfect balance between operational efficiency, passenger satisfaction, and sustainability. By strategically aligning priorities and leveraging AI, airports can optimise operations, enhance on-time performance, and delight customers.
The intelligent airport of the future is closer than you think.
Veovo’s “Airport ML at scale” approach leverages insights from airport client engagements, resulting in a prediction engine that not only enhances our product range but can also be seamlessly integrated as a service into existing infrastructure. We believe this represents the future of airport technology, where advanced services seamlessly integrate with operational applications like Resource Management Systems to deliver real operational benefits.
Get in touch to learn more about moving beyond traditional resource management and how you could take steps towards Total Airport management.
