Abilities and limitations in counteracting urban heat islands

Abstract

The article presents observational and simulation trends that have been used to mitigate the urban heat island phenomenon. Due to extensive progress in computing, researchers have focused on methods based on artificial intelligence. AI offers significant potential in counteracting UHIs through rapid temperature prediction, high-resolution mapping, optimisation of green spaces, and decision support. While AI models can expedite predictions, their precision may still lag behind traditional numerical weather prediction methods. The presented tools have major limitations, including the complexity of applying artificial intelligence to process the large datasets generated during city operations. The article presents the advantages of using AI algorithms to mitigate urban heat islands, which include more efficient management of energy, heating, and water infrastructure, as well as support in urban analysis. The ability to process vast amounts of data allows for more effective use of energy and water resources in buildings while continuously improving the algorithm. The literature review involved a comparison of current methods used to counteract urban heat islands, as well as existing and potential applications of artificial intelligence algorithms described in the Scopus and Web of Science databases. The analysis included the classification of the main groups of AI algorithms, their capabilities and limitations in mitigating urban heat islands, and an overview of the main trends in contemporary scientific articles on the use of AI in architecture. The selection criteria for the literature included the publication period (2016–2024), the applicability of the described solutions, and the current limitations and disadvantages of their use. The article highlights the main challenges in the widespread adoption of artificial intelligence in architecture.