Automated Deviation Analysis for As-Built Status Assessment of Steel Assemblies and Pipe Spools

Abstract

Steel assemblies and pipe spools play an essential role in the industrial construction sector. Fabrication of steel assemblies has been a challenging task due to the limited fabrication precision of the tools used in the process and inadequate inspection during fabrication. Moreover, unfavorable deformations may occur during the transportation phase which makes the erection and installation phase more complicated. These deviations require further considerations for realignment and repair that are associated with rework on construction sites. Hence, a systematic and automatic framework is required to continuously monitor the fabrication and installation processes of steel assemblies. Current approaches lack a sufficient level of control and are prone to error. This paper presents an automated framework to detect defective parts in steel assemblies and pipe spools in particular. A laser-based point cloud, which represents the as-built status, is compared to the original state from the CAD drawings that exist in the Building Information Model (BIM). Therefore, the defective parts are detected in a timely manner. The comparison is distance based and the procedure is fully automated. The experiments conducted to validate the proposed approach show that the model has high precision and a high rate of recall and has the potential to be employed for automated damage detection in order to improve productivity on construction sites. PROBLEM STATEMENT Steel structural assemblies are one of the critical components in both residential and industrial construction. Pipe spools are also key components for most industrial construction projects such as refineries and power plants. Utilizing steel structural assemblies on construction sites requires accurate fabrication and incident free shipment to site. Due to the advantages that staged fabrication of modules provides such as a controlled environment for fabrication, improved safety and improved

Original document

The different versions of the original document can be found in:

• http://pdfs.semanticscholar.org/7ddd/71321f21734cdc35d8e6ff87083528992661.pdf

• http://ascelibrary.org/doi/pdf/10.1061/9780784413616.256,

http://dx.doi.org/10.1061/9780784413616.256

• https://ascelibrary.org/doi/abs/10.1061/9780784413616.256,

https://academic.microsoft.com/#/detail/2116143626