Reclaimed Design: An Availability-Oriented Design Methodology for Reclaimed Lumber
Clara Blum, Laura Marsillo, Gonzalo Muñoz Guerrero
Supervisor: Prof. Thomas Wortmann, Prof. Jan Knippers
Advisors: Zuardin Akbar, Tzu-Ying Chen
In an era of resource scarcity and climate change, this study emphasizes a shift towards resource-oriented design methodologies. A computational and fabrication framework for reclaimed lumber is developed to inform the structural design based on material properties, aiming for desired performance and design flexibility.
The challenge in designing with reclaimed lumber lies in the material's inherent variability, which is due to its diverse origins, irregular geometries, and differing prior uses. This research addresses the critical gap of integrating data from pre-processed reclaimed lumber within the construction industry by proposing a design method that considers the complexity of reclaimed materials. We developed a method that integrates material analysis, dataset generation, machine learning (ML), equilibrium-based structural design, fixed robotics, and subtractive manufacturing. Material datasets from physical samples are used to train ML models to predict compressive strength from a non-destructive test to create a high-resolution material gradient within the lumber piece. This enhances the structural design and implements a physical structure using advanced fabrication techniques.
This method aims to maximize reclaimed lumber utilisation, promote sustainable construction practices, and advance the circular economy in construction with wood.
Karolin Tampe-Mai
Dipl.-Ing.Graduate School & Early Career
Equal Opportunity & Diversity (provisional)
[Image: ©IntCDC]