Generative AI and Retrieval-Augmented Generation (RAG) in an Agent-Based Simulation Framework for Urban Planning
A conceptual framework that integrates Generative AI and Retrieval-Augmented Generation with the principles of strategic engineering to enhance urban planning simulations and promises to transform urban planning into a more adaptive, inclusive and resilient practice.
Abstract
Urban planning is a multifaceted discipline that requires balancing economic growth, environmental sustainability and community needs. Traditional approaches often rely on static data and manual analyses, which can be time-consuming and less responsive to real-time changes. This paper proposes a conceptual framework that integrates Generative AI and Retrieval-Augmented Generation (RAG) with the principles of strategic engineering to enhance urban planning simulations. By leveraging real-time data and advanced modeling and simulation capabilities, this framework addresses the complexity inherent in urban systems. Generative AI, exemplified by models such as GPT-4, excels at producing coherent and contextually relevant text, while RAG ensures the incorporation of up-to-date, domain-specific information. The framework employs autonomous agents within the simulation software to dynamically model various urban development scenarios, providing planners with actionable insights that promote sustainability. The proposed system enhances decision-making, operational efficiency and community engagement by offering real-time, data-driven insights. Furthermore, it aligns urban development projects with long-term sustainability goals, fostering transparency and public trust. This interdisciplinary approach, rooted in strategic engineering, promises to transform urban planning into a more adaptive, inclusive and resilient practice.