Machine Learning to Predict Convective Weather and its Impact on En-Route Capacity
Paper ID
Conference
Year
Theme
Project Name
Keywords:
Authors
DOI
Project Number
Abstract
Network performance is very sensitive to weather and uncertainty in its prediction. We address these challenges through the contribution to an Artificial Intelligence (AI)- based Network Operations Plan. This plan is enhanced by including a probabilistic weather prediction tailored to ATFM, ATM and weather data integration and demand and capacity imbalance characterization at the pre-tactical and tactical phases of ATFM. We integrate all these modules into a visualization tool aimed at supporting human’s decision-making. An operational assessment has been conducted. The improvements for FMPs and NM in pre-tactical and tactical ATFM are: 1) Ability to understand the weather convective prediction and to identify critical weather areas; 2) Ability to understand and manage the capacity reduction prediction; 3) Ability to manage the situation at the network level; 4) Ability to improve the efficiency and productivity of human performance (workload, usability).