Master of Science in Engineering (MSE)
Civil and Environmental Engineering and Construction
First Committee Member
Jin Ouk Choi
Second Committee Member
Pramen P. Shrestha
Third Committee Member
Jee Woong Park
Fourth Committee Member
Number of Pages
As-built documentation of modular structures is an important set of records for a project, that consist of construction drawings and specifications, project design modifications, prefabrication, and component assembly. In particular, manual geometric quality assessments of a structure’s prefabricated components are generally fraught with errors, making it extremely difficult to compare the as-built construction with the specified drawings, which affects the project. Furthermore, if the prefabricated components are manufactured with errors, this can result in construction schedule delays and high additional costs due to rework. Several technologies, such as point cloud data generated using laser sensors and LiDAR, have been developed to verify as-built construction accuracy. However, such technologies have been associated with high equipment costs, equipment mobilization difficulties, high computing powers, long-duration to generate models, and the need for expertise. The photogrammetric approach for point cloud generation has advantages in terms of cost, easiness of data collection, and shorter time to generate models. However, due to a lack of capabilities to generate high-quality, accurate point clouds, there were minimal research studies. However, there was an advancement in the photogrammetric approach in terms of software recently. This research aims to verify an advancement of the photogrammetric approach for generating a 3D point cloud model of an existing structure, especially in modular construction, using pictures taken by a digital handheld camera, followed by refinement of the model, and an accuracy assessment, compared to the 3D BIM model. The proposed approach consists of taking photographs of the structure at equidistant viewpoints around the structure and processing the images for the point cloud generation. Furthermore, geometric quality assessment was conducted by comparing point clouds’ dimensions with 3D BIM model dimensions to analyze the dimensional accuracy of the point clouds. This research demonstrated the development of an accurate point cloud model of a modular house with a photogrammetric approach through numerous trials and errors. The error percentage of the best model was a range of -0.6133% to 1.0514% for structural geometry and 0.7250% to 0.0011% for building components. Additionally, challenges faced during this research and reasons for model generation failure were analyzed, and the lessons learned from it were documented to establish a foundation for future research. This study demonstrated that using a photogrammetric approach to develop point cloud models in the modular construction industry can be useful for geometric quality inspection of the structural elements, and several other purposes. The study also revealed that photogrammetric point cloud models could be generated with high accuracy, at a low cost with a reasonable time and effort.
Dimensional analysis; Geometric quality; Modular construction; Modular house; Photogrammetry; Point clouds
University of Nevada, Las Vegas
Kapsikar, Preetam, "Geometric Quality Assessment of a Modular House using BIM and Photogrammetric Point Clouds" (2021). UNLV Theses, Dissertations, Professional Papers, and Capstones. 4157.
IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/