The course begins with a foundational introduction to graph theory, focusing on how nodes and edges can model spatial, functional, and semantic relationships within building data. Participants will then learn how to convert parametric models from Grasshopper into the Industry Foundation Classes (IFC) format—a widely adopted open standard for building information modeling. The next step involves transforming IFC data into graph structures that capture the hierarchical, relational, and geometric information inherent in a building model.
MaCAD Digital Tools for GRAPH MACHINE LEARNING SEMINAR
Graph Machine Learning is an in-depth course that explores the powerful potential of representing buildings and built environments as graph structures. As the complexity of architectural and construction data increases, traditional methods of storing and analyzing this information often fall short. Graphs offer a compelling alternative—providing a way to model not just the elements of a building, but the rich web of relationships between spaces, components, systems, and behaviors. This course is designed to empower participants with the skills to convert design data into structured graph representations and leverage state-of-the-art tools to extract insight, perform advanced analysis, and enable intelligent applications.
The course begins with a foundational introduction to graph theory, focusing on how nodes and edges can model spatial, functional, and semantic relationships within building data. Participants will then learn how to convert parametric models from Grasshopper into the Industry Foundation Classes (IFC) format—a widely adopted open standard for building information modeling. The next step involves transforming IFC data into graph structures that capture the hierarchical, relational, and geometric information inherent in a building model.
Once the data is structured as a graph, the course transitions to the practical use of Neo4j, a leading graph database platform. Participants will learn how to import and organize their data in Neo4j and use Cypher, its declarative query language, to explore, analyze, and visualize the building graph in GraphML I. In GraphML II, the course delves into advanced topics, including the use of the Neo4j Graph Data Science (GDS) library for machine learning tasks such as clustering, link prediction, and node classification. The course concludes with an introduction to GraphRAG (Graph Retrieval-Augmented Generation), where participants will see how graph databases can be integrated with large language models to create intelligent, context-aware systems for querying and reasoning over building data.
By the end of the course, participants will have a practical, end-to-end understanding of how to model, analyze, and enrich building information using graph-based techniques—unlocking new possibilities for design intelligence, digital twins, automation, and AI-enhanced workflows in the built environment.
Using Node Similarity and Machine Learning: How advanced graph analytics and spatial daylight autonomy analysis are transforming architectural design workflow Introduction In an age where climate-driven architecture is paramount, daylight performance remains central to both spatial experience and energy efficiency. Our project tackles this by combining IFC data, graph logic, and machine intelligence to create … Read more
Nestled within IAAC’s Master of Advanced Computational Design program, our recent investigation of Graph Machine Learning (Graph ML) sought to reimagine how we perceive, evaluate, and improve architectural spaces. Our team of Andrea, Leila, Lennart, and Mahnoor worked on a courtyard prototype to create a seamless process that combines parametric modeling, semantic data interchange, and … Read more
This project introduces a graph-based machine learning system designed to automate furniture placement in apartment layouts. The workflow was developed as part of an architectural design studio and played a key role in generating comfort-driven interior configurations. At its core, the system encodes spatial and functional rules such as placing a sofa near a window … Read more
In modern residential architecture, acoustic comfort is as vital as daylighting or thermal performance. In our latest studio research at the Institute for Advanced Architecture of Catalonia, Group 09 tackled the challenge of mapping and optimizing sound behavior in homes using graph-based machine-learning workflows. Over the next five minutes, you’ll discover how we combined city-scale … Read more
Abstract In this project, we explore how graph-based thinking can reshape construction planning by bridging design data and scheduling logic. Drawing inspiration from modular architecture and network theory, we investigate new ways to visualize, analyze, and optimize the sequencing of building elements. By combining insights from BIM, parametric modeling, and graph analysis, the work aims … Read more
Our Solution: The Co-creator App We’ve developed a Copilot App – a local AI system that lives in your building and speaks your language. Literally. You make natural language requests, and it translates them into real architectural actions by understanding the complex relationships between spaces and residents through graph databases and GDS algorithms. Here’s how … Read more
In this project, we leverage Graph Machine Learning to enhance fire safety analysis by predicting fire spread risk in building units using enriched IFC models and advanced graph algorithms. Idea & Goal Our workflow begins in Grasshopper, where we configure building units and enrich the IFC model with custom fire-related attributes such as Material combustibility, … Read more
GWP-data enriched graphs Imagine being able to see the carbon footprint of a building, not after it’s built, but while it’s still a sketch. What if architects and engineers could get real-time sustainability feedback the moment they decide on a material or tweak a wall layout? That’s the vision behind Visual GWP. We combined Graph … Read more
