As Iceland’s main gateway, Keflavik International Airport (KEF) wanted to ensure passengers always enjoyed a seamless curb-to-gate experience, no matter their travel season. So, it turned to dynamic forecasting – allowing the airport to adjust capacity planning when conditions change rapidly.

Agile, accurate planning

Keeping passengers flowing smoothly through the airport, no matter what time of the year, is a key priority for KEF. To flex to changing conditions, the airport knew it needed to keep its planning window as short as possible. And that required harnessing live information.

Having experienced success with Veovo’s Passenger Flow technology in gaining airport-wide passenger flow insights, KEF decided to add Passenger Forecasting to its current system, allowing it to rapidly adjust capacity to demand, improve journeys and keep costs down.

Responding to unforeseen events faster

Since going live with the new solution, KEF has seen immediate improvements in performance and efficiency, with better data visibility allowing for more accurate planning. Veovo’sForecasting capabilities help the airport accurately predict passenger show-up profiles, allowing it to easily adjust resources to prevent bottlenecks and crowding and keep passengers moving.

The machine learning-powered solution draws from multiple sources across the airport, including various sensor technologies, AODB and boarding pass scans to create an accurate by-flight prediction of how passengers will move through the terminal in the days and weeks ahead. This enables KEF to build a data-driven, robust capacity plan. As the solution continually re-forecasts movement using live data, the airport gets an early heads-up of likely deviations from the plan in the days and hours ahead. This allows it to take proactive action, such as postponing breaks, reconfiguring lanes or prolonging a shift.

Building trust. Powering up performance

Previously, evaluations called for a lot of manual work, starting with collecting data from multiple systems and collating it into an Excel file. Now, the data is automatically accessible and presented visually per flight, allowing KEF to compare forecasts against the actual situation and easily calculate the accuracy of the forecasts.

It also allows KEF to understand the cause of a deviation from the plan, which could be anything from a late arrival to a slower-than-predicted security process. This insight is then fed back into the plan to continually improve forecast accuracy based on local conditions.

Test, compare, adjust, learn

With dynamic forecasting, KEF can test new metrics quickly, assume process times and compare these to plans that use different assumptions. It can easily adjust its forecasts by adding an unforeseen event to the calendar – for example, the implementation of new travel restrictions to specific destinations – which the system recognises and treats as unusual and unique.

Another advantage is that the solution is automatic – requiring no manual work or any specialised skill to use. All forecast changes or adjustments that KEF makes are easily managed in the system, leaving more time for evaluation and adjustments to improve future plans.

Thordis Bjørnsdottir Specialist in Operation Research, Keflavik International Airport commented

“Live re-forecasting is the most outstanding feature of the system. It continuously takes into account all variables of the current situation – such as queue length, scheduling, actual passenger counts and wait times – and compares them to the daily forecast. It then provides an updated passenger forecast and capacity plan every few minutes. By simply understanding how the actual situation deviates from expectation, we can almost instantly see the potential effects and mitigate these.”