Spatial Parameterization of Non-Semantic CAD Elements for Supporting Automated Disassembly Planning

  • Chris Rausch Department of Civil and Environmental Engineering, University of Waterloo
  • Ben Sanchez Department of Civil and Environmental Engineering, University of Waterloo
  • Carl Haas Department of Civil and Environmental Engineering, University of Waterloo

Abstract

Digital data and associated semantics play a fundamental role in supporting the vision of Construction 4.0. Advancements in digitization workflows such as scan-to-BIM and automated meta-data generation are being used for data-driven decision making. A challenge with collecting and processing raw, non-semantic data is the process of integrating intelligence into and characterizing data automatically. This paper demonstrates how spatial parameterization (i.e., extracting, modifying and analysing parameters that define the spatial properties of a component) can be used as a method for automating steps in disassembly planning for buildings. The potential use cases of disassembly planning include adaptive building reuse, robotic assembly programming, reconfigurable prefabricated assemblies and selective disassembly for rehabilitation and repairs. This paper presents spatial parameterization in a framework to disassemble building components via a rule-based algorithm that comprises three dimensional Cartesian properties and clash detection between non-semantic CAD elements. Demonstration of the framework is carried out using a case study where the interior wall of a building on the University of Waterloo campus was disassembled for adaptive reuse purposes. Comparison of the case study results to the actual disassembly sequence demonstrates how spatial parameterization is effective for automating key steps in disassembly planning. A discussion is provided to identify key barriers to increased automation which relate to modelling accuracy, Level of Development (LOD) for Building Information Modelling (BIM), and global spatial constraints for disassembly.
Published
2019-05-24