SMART BRICK

Computational Design
Smart Brick explores the production of specialized building components in an experimental manner. We developed a computational strategy for designing construction elements. We investigated a range of different mathematical models of computational design for creating and modifying three-dimensional, fractal-like tree structures, which topologies are not limited by predefined standards or protocols. Intricate internal lattice structures allow architects to tailor the porosity, reduce weight, or tune structural performance properties.

Our models were also designed to meet fabrication conditions and requirement. Because sand-printed elements can only take compression forces, the branch-like components are designed so that, after pre-stressing, they exhibit the superior capacity for loading compression. Through several iterations joints we designed that interlock the components, we then assembled the branch-like components as one fixed tree-like structure, which we regard as one structural element. Designing 3D printed building components stimulates the creation of a computational design system that, compared with conventional software, generates more complex printable geometries with higher resolutions. Such a system also has the benefit of integrating mass customization (to be adjusted to specific requirements) into the design and fabrication process.

Fractal Geometry and Tree Structure Model
A branch of mathematics studies abstract configurations characterized by patterns of self-similarity and recursive growth. The development of ‘graphic statics’ as a theoretical approach, helps to understand the association between structural forms and equilibrium of forces.

With such knowledge, architects abstracted the form of tree's fractal configuration to build dendriform structures like a mushroom or an umbrella during the early and mid-20th century. In the past, several technical ways were exercised to connect fractal concepts with architecture by physical modelling process. Nowadays, with the advantages of contemporary computer technology, a procedural generative approach becomes possible based on compositions of mathematical functions.  

Project Credits: MoonYoung Jeong, ZongRu WU.
This project was developed during one week called Smart Brick in ETH Zürich.
Supervisors: Mathias Bernhard, Patrick Bedarf.
Photo Credits: Tom Mundy.