ABSTRACT
This paper explores deploying a custom computational tool for segmenting point clouds in the context of a circular economy and the reuse of vacant buildings within a multinational architectural practice spanning Denmark, Sweden, and Norway. Focused on early-phase reuse analysis, the tool ReUseX leverages LiDAR scans and machine learning algorithms to automate the generation of semantically classified point clouds and solid surface models. The primary challenge lies in ensuring accessibility across different software platforms while maintaining the tool with a small development team. Through workshops and field studies, insights into user experiences highlight the importance of speed, ease of use, and interoperability. The paper discusses strategies such as utilizing Speckle as an intermediary layer for data exchange and proposes enhancements like developing a custom scanning app and incorporating voice annotation for metadata. This contribution offers practical insights for integrating computational tools in architectural practices, advancing sustainable building reuse practices.
