Research

This research area focuses on identifying, documenting and assessing existing building materials to enable reuse. Work in this domain explores how computer vision, natural language processing, deep learning, and reality capture methods can automate building audits and generate reliable digital inventories. The aim is to reduce uncertainty around material type, geometry, condition and reuse potential, thereby supporting early circular decision making. Research also investigates how digital workflows can convert raw visual and spatial data into structured models that feed into BIM systems and digital material banks.

This research area investigates how selective deconstruction can be carried out with greater safety, precision, and resource efficiency. The work focuses on robotic systems, sensor integration, and control strategies using machine learning to enable structures to be systematically deconstructed for reuse rather than demolished. Such approaches aim to preserve the value of building components while reducing risks for workers and maintaining structural stability throughout staged disassembly. Research also addresses the uncertainty inherent in existing buildings and construction sites. Robotic workflows therefore need to remain adaptive, capable of responding to irregular components, incomplete information, and changing structural conditions during the deconstruction process.

The distribution phase examines how reclaimed materials are documented, tracked, matched, and circulated between projects. Research in this area develops digital passports, data standards, and automated data-processing pipelines that organize and maintain reuse inventories. Additional work studies digital material banks and track-and-trace systems that preserve the identity and history of materials across multiple life cycles. Research also investigates digital platforms that operate as procurement infrastructure for reuse. These platforms support matchmaking between available reclaimed components and project requirements, while integrating considerations such as timing, logistics, risk, and stakeholder coordination. The objective is to make reuse transactions more reliable and operational within construction practice.

Research within the design category examines how reclaimed materials can actively shape architectural design processes. One line of work develops co-design tools that use machine learning to help designers assemble heterogeneous reused components into structurally and spatially coherent configurations. These tools complement human judgement by proposing design alternatives that respond to variability in material availability, geometry, and performance constraints. Another line of work studies the use of language models to support life cycle assessment workflows. These systems assist with tasks such as emissions factor matching, process inventory generation, and the comparison of multiple environmental scenarios in accordance with established standards. Together, these approaches aim to make circular design decisions more informed, efficient, and performance driven.