Optimization of Airport Resources through Air Traffic Prediction
The entity responsible for air navigation (through its CRIDA laboratory) needed to anticipate airport traffic volumes more accurately to optimize the allocation of its operational resources, such as air traffic controller shifts and the provision of essential airport services. Although they had aggregate-level forecasting systems, they required a much more granular approach that would allow them to plan with high reliability over time horizons extending up to four months.
Within the framework of the Enaire Open Innovation program, an advanced analytics solution focused on individualized flight-by-flight forecasting was developed as an alternative to aggregate predictions. The system ingests two distinct sources of airport planning data and cross-references them with a third historical data source using an automatic classification model. This machine learning algorithm evaluates and identifies the actual probability of each planned flight taking place in reality, and allows for the introduction of potential new candidate flights, making it possible to reconstruct the global aggregate forecast from the finest data point available.
This model gives planning teams an extremely precise view of the real operational load that airports will bear in the medium term. By predicting traffic from the highly granular individual level of each flight rather than a broad aggregate, the entity can optimally size its work teams. This guarantees total coverage of necessary services during peak demand hours and maximizes the profitability and efficiency of human resources during periods of lower activity.
.jpg)
Automated Legal Watch & Alert Generation with AI
Automated AI Tender Generation
