Automated verification of 3D manufacturability for steel frame assemblies

Shi An, Pablo Martinez Rodriguez, Mohamed Al-Hussein, Rafiq Ahmad*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)

Abstract

The emergence of off-site construction has shifted many construction activities from job sites to controlled factory environments, enabling the extensive use of automated machines in the fabrication of construction-oriented products (such as fame assemblies). When using highly automated machines in agile manufacturing, the process planning becomes the bottleneck since it is mostly manual, experienced-based, and error prone. Even though the manufacturability was proven to be automatically evaluated for 2D wood-framed assemblies, many construction-oriented products, such as steel-framed assemblies, have three-dimensional features that have not yet been accounted for. This paper extends the previously developed 2D framework for 3D applications as required in steel-framed assemblies, in order to automatically check the manufacturability of steel frame assemblies given the building information model (BIM). The proposed system detects intersection regions in a frame assembly and calculates areas that require manufacturing operations, such as fastening with screws. This objective is accomplished with the use of classic techniques commonly used in computational geometry. The proposed framework is validated using a steel frame with commonly encountered geometric features in the machine environment. The result proves that the proposed approach automatically and accurately determines the manufacturing locations of the frame assembly.
Original languageEnglish
Article number103287
Number of pages14
JournalAutomation in Construction
Volume118
Early online date5 Jun 2020
DOIs
Publication statusPublished - 1 Oct 2020
Externally publishedYes

Fingerprint

Dive into the research topics of 'Automated verification of 3D manufacturability for steel frame assemblies'. Together they form a unique fingerprint.

Cite this