<p>Convective weather&#160;represents a significant disruption to air traffic flow management (ATFM) operations. Thunderstorms&#160;are the cause for a substantial amount of delay in&#160;both the en-route and airport&#160;environment. Before the day of operations, poor prediction capability of convective weather prohibits traffic managers from considering weather mitigation strategies during the pre-tactical phase of ATFM planning. As a result, convective weather is mitigated tactically, possibly leading to excessive delays. &#160;</p><p>The skill of weather forecasting has greatly improved in recent years. Hi-resolution weather models can predict the future state of the atmosphere for some weather parameters. However, incorporating the output from these sophisticated weather products into an ATFM solution that provides easily interpreted information by the air traffic managers remains a challenge.&#160;</p><p>This paper combines data from high-resolution&#160;numerical&#160;weather&#160;predictions&#160;with actual storm observations from lightning detecting and satellite images. It applies supervised machine learning techniques such as binary classification, multiclass classification, and regression to train neural networks to predict the occurrence, severity, and altitude of thunderstorms. The model predictions are given up to 36hr in advance, within timeframes necessary for pre-tactical planning of ATFM, providing traffic managers with valuable information for developing weather mitigation plans.&#160;</p>
<p>Flight delays are one of the major concerns in air traffic management. The impact of flight delays represents financial and time losses and may derive in loss of reputation of the air traffic business. On average weather accounts for roughly one-third of ATFM (Air Traffic Flow Management) delays (25% for en-route and ~50% for the airports). Examining only the top ten days with highest delays due to weather regulations from the first half of 2018, strong convective activity throughout Europe was the principal cause, with estimated cost due to airport and en-route delays reaching almost &#8364;130 million (roughly 10% of the weather delay in 2018 concentrated in only 10 days). Given these large cost figures, even minor improvement in prediction and performance of ATFM operations during significant convective weather events will yield to substantial yearly savings for the ATM (Air Traffic Management) system. Designing an efficient value chain for ATFM, that propagates weather forecasts into a series of tools to select mitigation measures at local and network levels in a collaborative ATFM operations paradigm, requires a multidisciplinary approach to gather the different stakeholders. Such an approach has been developed in the SESAR ISOBAR project, whose aim is to integrate enhanced convective weather forecasts for predicting imbalances between air traffic capacity and demand (requests to fly by airspace users, mainly airlines) and to select appropriate mitigation measures. The value chain developed in the framework of ISOBAR leverages the power of Artificial Intelligence (AI) in the different stages. AI engine is trained using a dataset of selected convective events in summer 2019, which includes forecasts from high resolution ensemble prediction systems (IFS, &#947;-SREPS and AROME-EPS), declared capacity in air traffic flow and initial air traffic demand. The value chain produces a solution for tactical (day 0) and pre-tactical (day -1) ATFM operations. A validation exercise was organised at EUROCONTROL Innovation Hub in March 2022 with the collaboration of ATC (Air Traffic Control) operational staff and Air Traffic Controllers from Spain, France and Europe air traffic network management. AI Engine was run for some high convective situations over Europe, which were characterised by high delays due to weather regulations. Offline simulations highlighted the added value of the solutions assessed by ATC experts. The predicted air traffic delay has been drastically cut by up 75%.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.