ARCHITECTURAL DESIGN, BIM, AND ARTIFICIAL INTELLIGENCE: A REVIEW OF RULE-BASED AND MACHINE LEARNING APPLICATIONS

Abstract

Architectural design projects can reach high complexity, long lists of requirements and intricate architectural programs, require effective ways to assess and predict building performance. Building Information Modeling (BIM) has proven to be a valuable tool in improving the design process by offering a range of resources for managing geometric data and building performance. However, BIM's current limitations, particularly in terms of geometry and function representation and its ability to predict performance, hinder its full potential. In this context, Artificial Intelligence (AI) is presented as an opportunity to enhance BIM workflows. AI's capabilities in prediction, automation, and analysis have the potential to improve design processes, offering more accurate simulations and predictions. This research explores how AI can be integrated into BIM-supported architectural design by conducting a case study using early-stage design AI tools, specifically Autodesk Forma. The study examines how AI-powered design tools can assist in the creation of building designs that meet specific performance criteria, such as energy efficiency, sustainability, and compliance with local regulations. Through this case study, the research aims to provide insights into how AI is helping to refine the BIM design process and to uncover the potential of AI in overcoming BIM's limitations. Ultimately, the findings will highlight the growing role of AI in architecture and offer perspectives on its future integration into design workflows.