Lay Summary

Global challenges like urban energy efficiency and natural disasters require rapid, data-driven decisions. This paper presents two real-world case studies demonstrating how artificial intelligence can tackle these complex issues. First, deep learning models predict urban solar radiation coverage to help plan renewable energy installations in cities. Second, machine learning algorithms analyze environmental data to forecast wildfire spread and support emergency management. By integrating AI with high-performance simulations, the researchers demonstrate how modern computational methods can enhance climate resilience and disaster response. (84 words)

Full Abstract

Global challenges such as climate change, natural disasters, and energy transitions require innovative computational methodologies to deliver fast and reliable solutions. This paper addresses these gaps by employing Artificial Intelligence (AI) and High-Performance Data Analytics (HPDA) to enhance predictive accuracy and data handling in two critical areas: predicting shading effects between buildings for sustainable urban planning, and improving wildfire management through pre-computed simulations and deep learning models. Our approach utilizes neural networks to model urban solar exposure accurately and leverages HPDA to process extensive wildfire data for better preventive measures and response strategies. The findings show that integrating AI with HPC infrastructure yields substantial scalability and accuracy gains across environmental simulation contexts.

Metadata

• Publication Date: November 2024
• Author: Giorgos Filandrianos, Angeliki Dimitriou, Vasiliki Kostoula, Nikolaos Chalvantzis, David Caballero, Luis Torres, Michal Kulczewski, Javier Cladellas, Zoltán Horváth, Harald Köstler, Konstantinos Nikas, Dimitrios Tsoumakos, Giorgos Stamou
• Conference: AISyS 2024 (First International Conference on AI-based Systems and Services)