USF Inventors have developed Deep Learning Framework that uses high-precision gridded weather forecast data by two novel contributions. One: it uses assembled gridded weather forecast for the terminal airspace instead of an isolated station-based terminal weather forecast that has been used in existing literature. Second: it proposes a multi-layer convolutional neural network to predict both runway configurations and AARs for different airports in a multi-airport system. This invention developed an end-to-end system to predict real-time runway configurations and AAR simultaneously for multi-airport system, where synchronized air traffic operations are presented.
Typical Airline Production Process