Bio-Digital Intelligent Agents As A Tool For Flow-Driven Architecture: Interactive Workflow Enhancing Decision-Making At Early Design Stage

This paper explores the integration of Agent-Based Modeling (ABM) into architectural design, aiming to enhance early design stage through a flow-driven approach. It proposes a method where intelligent agents, mimicking process of natural morphogenesis, aid in data analysis and decision-making. The framework introduces sensing agents that explore site data, processing agents that distribute themselves based on collected observations and project data, and guiding agents enhanced by Large Language Models (LLM) that enable user interaction through textual commands. The system’s emergent output is a flow boundary, a mesh that visualizes agent movements and interactions with the environment, providing architects with a dynamic, real-time view of design performance. This tool enhances architects’ ability to communicate with data, promoting exploratory, collaborative, and interactive design process while offering flexibility in decision-making without diminishing the architect’s control.