BIM Scan

[1]

3D laser scanning is already a permanent element of BIM inventory. As part of the implementation of various projects, designers encounter countless barriers and new opportunities offered by this technology. In the article - BIM Scanning Methodology Used in Heritage Buildings [2] - Gustavo Rocha OrcID, Luís Mateus, Jorge Fernández and Victor Ferreira from 2020, you can read more about the implemented processes and procedures. Below I present the most important fragments:

Historic buildings tend to have a complex (non-parametric) geometry that transforms their digitization by conventional methods into inaccurate and time-consuming processes. When it comes to studying and representing historical assets, remote sensing technologies have played a key role in the last few years: 3D laser scanning and photogrammetric measurements save time in the field while proving extremely accurate in capturing irregular building geometries. However, efficiently transforming remote sensing data into as-built intelligent parametric models is currently an unresolved challenge. A pragmatic and structured Historical Building Information Modeling (HBIM) methodology is essential to arrive at a coherent model that can benefit and integrate conservation and restoration work. This article is about building an HBIM heritage asset model using 3D laser scanning and photogrammetry. Our findings are illustrated by one case study: The Engine House Paços Reais in Lisbon. The article first describes how and what measures should be taken to schedule an in-depth scan to HBIM process. Second, the description of the remote sensing research campaign is carried out appropriately and is aimed at obtaining BIM results, including the process of matching, cleaning and fusing data. Finally, the HBIM modeling phase is described based on the point cloud data.

Cited Document

In Europe, 80% of buildings were built before 1990 and most of them do not have a BIM model that could be included in this working methodology. In such cases, reverse engineering with the use of 3D laser scanning and photogrammetry becomes a standard procedure [4]. When dealing with a historical building, care should be taken to arrive at an appropriate model that can meet the needs of the Historical Building Information Modeling (HBIM) methodology. It is common for the term BIM to be misused because it is often associated with software rather than a process. BIM is not software, but an integrated, collaborative methodology focused on a digital building model that provides the information needed to manage a building throughout its life cycle, from design and construction to maintenance and management after use.

The term HBIM is described by Murphy [11] as a parametric model generation solution in which architectural elements are represented not only in their geometry but also in the corresponding historical database attributes. It is the application of the BIM methodology to listed historic buildings and can be for condition monitoring, heritage management, preventive maintenance, intervention option analysis, maintenance and restoration planning, construction simulation, disaster preparedness and more.

Historic buildings often have walls with heterogeneous thicknesses, deviations and non-perpendicularity. Non-orthogonal walls hinder the work of HBIM, and attention should be paid to this aspect to decide what approach will be taken. The engine room did not show large deviations of the internal walls, which allowed for orthogonal modeling of these elements. However, its outer perimeter was not perfectly rectangular. One of the walls showed a difference of more than 5 cm in its actual position from the expected perpendicular angle. In this case, after verifying the significant deviation, it was decided to model the wall in its actual position, with an appropriate deviation to keep the geometric properties consistent with reality. The walls were made with the appropriate thicknesses with an accuracy of 1 cm. It was also decided to separate exterior and interior finishes, allowing at the end of the process to generate a schedule with the number of finishes associated with exterior or interior uses. The model was created to avoid conflicts and collisions between structural elements. The modeling was performed by one team who had full control of how it was to be done to avoid these conflicts. The interaction of floors, beams, walls and other elements has been carefully crafted to ensure a perfect intersection and to reflect the original building with high fidelity (Figure 9). Even so, at some points we had problems with conflict, for example between trusses and walls and roofs. This was due to the impossibility of approaching these points, egez which measurements of the highest areas of the building were less accurate.

Complex structural elements require accurate and timely modeling and analysis. This goal is driving scientists to look for automated solutions in generating BIM data. Over the past five years, the number of BIM automation publications has increased by approximately 400% (229 articles from 2014-2019 and 57 articles from 2009-2013, downloaded from SCOPUS using the BIM and automation keywords). However, there is a discrepancy between industry practitioners and researchers, especially with regard to the criteria that guide the implementation of BIM in construction [38,39]. This mainly affects the implementation / integration of automation in real non-modular contexts. Complex forms, numerous morphological and typological variables remain the main barriers to the standardization of automated solutions in BIM modeling. However, there is a good chance for an increase in processing possibilities, and new knowledge in the field of graphics and computer vision will bring reliable standard solutions in the near future. As part of automation for BIM, topography modeling has received a lot of attention, and scientific achievements have been successfully implemented in commercial solutions. The topography around the case study building had a few slopes and deformations, but manual modeling still meant overwork and inaccuracy. To avoid errors and save time, an external plugin installed in Revit called "Scan Terrain" was used. This plugin allows us to automatically create a topographic surface from a point cloud. Users can define the size of the cloud crop, the distance between the created points and the limit of the height of the points that will be on its surface. In this way, a topographic surface can be created in seconds, but it is nevertheless necessary to inspect and correct some points that may have been created incorrectly. The plug-in algorithm can identify horizontal surfaces and avoid vertical elements such as walls and furniture, but some elements such as steps, floors, and plants can still be confused with topography. The lack of regularity in the point cloud, the presence of areas that were not well captured by the scan, and vegetation up to half the height also interfered with the proper operation of the plug. Therefore, for best results, pre-cleaning of the point cloud is required, and manual adjustments are required after topography has been created. Nevertheless, this method proves to be very effective in providing a model of high quality and accuracy and saving modeling time.

The proposed workflow (Figure 16) (Table 1) was to create a BIM model that is prepared to increase the level of detail if needed in the future. Some elements were modeled in more detail, reaching LOD 350 (trusses, doors and windows), while others remained on LOD 300 (walls and floors). This was due to the inability to verify the materials constituting the core of some elements. Thus, their model was limited to dimensions, size, position, orientation and finishing materials, but without distinguishing between the inner layers. The final model, in addition to the geometric reconstruction of the building, includes all elements classified according to its purpose, including finishing materials and structural and support elements, if possible. The BIM model is not static; LOD can be increased by updating geometric and non-geometric information (such as physical and material properties, costs, manufacturers, composition, and more) at any time.

Sources:

[1] pixabay.com; author: MarkusChrist

[2] BIM Scanning Methodology Used in Heritage Buildings - Gustavo Rocha OrcID, Luís Mateus †, Jorge Fernández † and Victor Ferreira † CIAUD, Lisbon School of Architecture, University of Lisbon, 1349-063 Lisbon, Portugal * Author to whom correspondence should be addressed † These authors contributed equally to this work, Dziedzictwo 2020, https://www.google.com/url?q=https%3A%2F%2Fwww.mdpi.com%2F2571-9408%2F3%2F1%2F4 % 2Fpdf & sa = D & sntz = 1 & usg = AFQjCNHGuCtSTWefVujF_0fNNUqFFv37dg