CAREER: Generative Deep Learning for Post-Disaster Spatial Regeneration Planning

The objective of this Faculty Early Career Development Program (CAREER) project is to support research on how to integrate artificial intelligence (AI) into spatial planning methodologies and processes. This knowledge is particularly important to informing long-term disaster recovery. By exploring Generative deep learning (GenAI) based approaches, the project promotes economic vitality, housing affordability, and other planning objectives while ensuring their responsible use. The methodologies developed are applicable to various planning and disaster contexts. The outcomes can significantly enhance disaster management capabilities and contribute to workforce development in AI and urban planning. The research addresses key challenges in formulating, training, and calibrating GenAI models and in preparing for their anticipatory governance in spatial regeneration planning processes. Specifically, it develops novel algorithms to balance multiple planning objectives, consider the effects of connected areas, respond to incremental planning across spatial scales, enable transactive planning, and avoid repeating past spatial disadvantages. It creates augmented spatial datasets and makes them available for public use. The project also examines how planners perceive and interact with these GenAI models, establishing responsible ways to deploy them. Educational efforts are designed to innovate curricula and teaching methods in urban analytics programs by combining learner persona-tailored pedagogies, new authentic learning modules, an online course, and original curricular studies. Integrated activities have the potential to advance cyberinfrastructure for planning research and education and contribute to building more resilient cities. The project is jointly funded by Humans, Disasters, and the Built Environment Program and Human-Environment and Geographical Sciences Program. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.