The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences (May 2023)

MONITORING WORK PROGRESS OF DAM CONSTRUCTION BASED ON PHOTOGRAMMETRIC POINT CLOUDS AND BIM: PRACTICAL APPROACH TO TEACHING INDUSTRY

  • A. P. Arbad,
  • Z. N. Arifin,
  • N. Martina,
  • E. P. Nindya,
  • M. A. Nurfa

DOI
https://doi.org/10.5194/isprs-archives-XLVIII-5-W1-2023-1-2023
Journal volume & issue
Vol. XLVIII-5-W1-2023
pp. 1 – 5

Abstract

Read online

Industry 4.0 encompasses a range of ideas, technologies, and procedures that impact engineering education and other fields. Building information modeling (BIM) has become a widely adopted practice in Indonesia's construction industry, prompting polytechnic universities to incorporate BIM skills and concepts into their construction engineering and management degree programs. Nowdays, more than 50 dam construction works occurred in Indonesia from 2014-2024. Monitoring progress regularly is crucial to ensure a construction project's successful completion. This can be done by visually tracking the project's advancement, which enables the identification of any changes or deviations that occur. Accurate measurement data is essential for reliable progress management, as it allows for the prediction of future project success or failure. Image-based object detection is highly beneficial for retrieving site information and monitoring construction progress. To monitor the progress of a construction project, comparisons between the building's as-built and as-planned states are necessary. BIM is employed to establish the intended condition of a building by providing information regarding its geometric characteristics and construction schedule. In this paper, we present a novel method for producing a point cloud that accurately represents the dam building site using photogrammetry. To overcome the limitation of capturing photographs from every possible angle at a construction site, we employ a combination of structure from motion and control points. This approach allows us to generate a scaled point cloud in a standardized coordinate system. The building's as-built and as-planned states are then compared using this point cloud. Structure from Motion (SfM) disparity maps are fused to create dense point clouds, which are then compared to the goal state for updating 4D BIM. This makes it possible to identify missing building components. To validate the presence of building elements, a secondary examination is conducted using the positions in front of and behind the model planes constructed as per the original blueprint. The research paper explores potential approaches and presents empirical results obtained from an actual case study.