AI-DRIVEN CLIMATE RESILIENT DESIGN – PREDICTING AND ADAPTING INFRASTRUCTURE TO EXTREME WEATHER PATTERNS USING AI.

Abstract

The aim of this study was to incorporate the use of Artificial Intelligence (AI) as part of the climate resilient infrastructure design, where AI can assist in resolving climate change related extreme weather event related challenges. Vulnerabilities of infrastructure at present, the use of AI to predict extreme weather as well as the use of AI to develop new as well as existing infrastructure to address climate risks are the focus of the research. By interviewing civil engineers and AI professionals qualitatively, investigating of the key barriers in integrating the AI into the civil engineering work, including data availability, the lack of technical skill or costs, are identified. The finding is that Long Short-Term Memory (LSTM) Networks and Convolutional Neural Networks (CNNs) are AI models that have great potential in improving accuracy of weather forecasting and deriving actionable insights for planning infrastructure. To support the framework, they also mentioned AI driven tools for infrastructure design such as flood proof buildings and retrofitting drainage system as important applications. However, only barriers in data quality, as well as the necessity of an interdisciplinary collaboration between researchers working in AI, civil engineers, and policymakers were identified. To make the most of the role of AI in creating resilient infrastructure in the changing climate, it details needs to enhance data infrastructure, invest in AI education, and foster cross disciplinary communication. The findings help understand the role of AI in adaptation of infrastructure against climate change.