Net-Zero Structural Design Using Synthetic Data. Generative structural design for rapid cost and embodied carbon evaluation
Interdisciplinary exchange between the organizers and the workshop participants;
Masterclass style sessions, discussion sessions and presentations.
Working directly on the app development
collectively brainstorm with the wider AEC community, generating ideas and develop concepts that can be used to further improve the tool, to become a viable solution for architects and engineers alike to perform and evaluate early-stage structural design.
Required software: Visual Studio Code. Possibly Rhino/Grasshopper/Karamba3D
Required hardware: Laptop with Windows 10 or Later
Embodied carbon from building materials and construction accounts for 11% of the total global carbon emissions with a building’s superstructure being the most significant source. The building industry urgently needs to transition to carbon sequestering structural materials to meet UN climate targets. However, the time-intensive nature of structural design creates a significant bottleneck. In early design stages, sharp deadlines mean architects and engineers will likely stick to tried-and-tested, but environmentally harmful, options like steel and concrete. As a result, to speed-up adoption of more sustainable structural materials, like mass timber, early-stage structural design must be re-imagined.
We have developed a novel data-driven structural design process which leverages synthetic data generation, and machine learning predictions to meet this demand. This user-friendly design tool (https://co2-demo.app.bluefoam.io/ ) was the result of a research collaboration between Digital Blue Foam and Karamba3D in initiated at DigitalFUTURES 2022. The first prototype allows project stakeholders to quickly generate and compare structural design options and scenarios with respect to cost, embodied carbon and sequestered carbon.
The goal of the workshop is to collectively brainstorm with the wider AEC community, generating ideas and develop concepts that can be used to further improve the tool, to become a viable solution for architects and engineers alike to perform and evaluate early-stage structural design. The current ML model should be expanded to include larger datasets and more complex projects. Current limitations to the structural typologies selected also limit the potential of the tool, often taking a conservative. Through incorporating data from existing buildings or projects we can start to validate the machine learning model’s predictions and provide more accurate and relevant results. The participants of the workshop will learn how to generate and use synthetic structural data in machine learning based workflows.
Students and design professionals from AEC who are enthusiastic about methods to address the life cycle footprint of buildings in early stage design.
Number of participatns max.
Duration and Procedure
- Introduction and outline of two days
- Masterclass Session – Geometry & Structural Analysis
- Brainstorm & Work Session – Geometry & Initial Structural concepts
- Masterclass Session – ML & Datasets
- Brainstorm & Work Session – Data Collection and ML Datasets
- Intermediate presentation & discussion
- Masterclass Session – UX and Tool functionality
- Brainstorm & Work Session – Use case scenarios
- Collation of findings and investigations
- Final presentation & discussion