Analysis and management of structural deformations through parametric models and HBIM workflow in architectural heritage

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ Esta es la versión aceptada del artículo publicado en: This is an accepted manuscript of a paper published in: Journal of Building Engineering (2022): 30/1/2024 DOI: https://doi.org/10.1016/j.jobe.2021.103274 Copyright: © 2021 Elsevier Ltd. All rights reserved. El acceso a la versión publicada del artículo puede requerir la suscripción de la revista. Access to the published version may require subscription. “This is an Accepted Manuscript of an article published by Elsevier in [Journal of Building Engineering] on [2022], available at: https://doi.org/10.1016/j.jobe.2021.103274” 1 Analysis and management of structural deformations through parametric models and HBIM workflow in architectural heritage. Juan Moyano 1, Ignacio Gil-Arizón 1, Juan E. Nieto-Julián 1 and David Marín-García 1 1 Department of Graphical Expression and Building Engineering, Higher Technical School of Building Engineering, University of Seville. 4A Reina Mercedes Avenue, Seville 41012 and Spain; [email protected] (J.M.); [email protected] (I.G.A.); ([email protected] (J.E.N.J.); [email protected] (D) * Correspondence: [email protected] (J.M.) https://doi.org/10.1016/j.jobe.2021.103274 Abstract Maintaining the structure and ornaments of historical buildings is essential to preserve cultural heritage in any society. Thus, dating the state and the evolution of the elements requires a special treatment, including the application of both advanced numerical analysis and noninvasive data acquisition techniques. In the latter, making digital copies of geometries through 3D reconstruction models is of great interest to compare and analyse structural evolutionary data accurately. For this purpose, appropriate software with containers of information correlated to the parametric elements in a BIM environment should be used. However, it is imperative to advance from static to dynamic models to collect the structural transformations caused by both the pass of time and other factors. The methodology followed in this paper is based on the experimentation by creating digital twins. The portico of a courtyard in a historical building from the 18th century was used as test bed. Based on the bibliographical recommendations, the terrestrial laser scanner is applied as a technique to acquire accurate data. The point cloud is used as a referential auxiliary to survey the model in the BIM platform with the Revit software. To assess the quality of the model built and to analyse the structural deviations between the parametric model and the actual geometry, the Dynamo script is used. To validate the experimentation, structural deviations are measured using both the parametric model and the point cloud with CloudCompare, a software for data treatment. The results were very positive because the deviation between the data obtained by Dynamo© and CloudCompare in the most unfavourable construction unit was between 0.5 and 1.17 cm, so these techniques are highly appropriate to review visual records and to analyse structural deviations. This new approach presents a new gap in the 3D reconstruction to date and control architectural structures, particularly in historical buildings. Keywords: HBIM, terrestrial laser scanning, generative algorithms, structural systems reverse engineering, cultural management 1.- Introduction Preserving heritage is essential in the cultural life of today’s society. For this reason, structures and ornaments should be appropriately maintained throughout historical buildings’ life cycle. Modern buildings are today subjected to a strong regulatory quality control, unlike historical buildings. Thus, determining the structural conditions of these immovable assets requires a special treatment not just through an advanced numerical analysis [1], but through noninvasive techniques consolidated under the umbrella of the European Federation for Non- Destructive Testing (EFNDT). Assessment techniques include the Structure from Motion/ Multi- View-Stereo (SfM/MVS) and the Terrestrial Laser Scanner (TLS), which are data acquisition techniques that provide a copy of the geometric surface of the elements to be analysed. These 2 techniques are very interesting to be applied in heritage buildings as these buildings usually have singularities in comparison with the other buildings, such as the composition of their geometry. Comparing and analysing these geometries, which are generally complex, by using these techniques could be an effective and accurate tool to study structural units. The elements that could be identified are pathological symptoms, cracks, collapse in walls and columns, deflections of horizontal elements, and detachments of stone elements, among others. Therefore, the use of the reverse engineering through SfM and TLS could be a nondestructive technique of great interest to assess and analyse the deformations of structural elements. In the diagnosis phase in which the building’s health is assessed, the geometry could be analysed as a pattern method. Jiang et al. [2] reviewed the use of the short-range photogrammetry in both the deformation of bridges and the measurement of their geometry. Xu et al. [1] used TLS for the static and dynamic analysis of the structural deformations of arches, concluding that placing sensors in structural elements is not required with this technique, unlike other procedures. The analysis is not just organoleptic though visual inspection tasks that could interpret structural damages; it is complemented with other analyses. Some methodological studies are focused on immersive environments between digital models and teledetection to diagnose the preservation of cultural heritage (CH) [3]. Today, architecture and construction (AC) in the 3D reconstruction model is fundamental to preserve CH. The geometry involved is usually complex, so the CAD/BIM software should be appropriately selected, and effective modelling tools should be used to be approached as accurate as possible, also considering time. BIM digital platforms represent a new paradigm both in the Architecture, Engineering and Construction (AEC) industry and the academic and scientific area. These 3D digital reconstructions are also containers of information correlated to the parametric elements. However, the construction model is a theoretical model built by the operator by using primitive geometries, i.e., drawings [4] or point clouds [5,6]. This model does not consider structural transformations due to the pass of time or those caused by natural circumstances (e.g., earthquakes and hurricanes), among other external agents. For this reason, there is a knowledge gap related to the assessment in various periods of the quality of the model built, so the structural deviations between the theoretical model and Massive Data Capture Systems (MDCSs) should be analysed. This process presents a new approach in architectural structures and determines processes to consolidate 3D records that could be inherited by future generations, developing technologies that ease complex analytic processes. The contribution of this paper is therefore focused on the codification of an algorithm within the applicability of the BIM digital platform, which analyses the structural deformations of buildings. It is initially applied to historical buildings, but it could also be applied to new buildings. Inferring an algorithm from the digital platform to analyse the deviations between MDCSs and (ideal) parametric objects means to assess in real time both the quality of the model built and the structural deformations that could take place throughout the life cycle of the building. This process simplifies the use of other complementary software within the structure scope, as well as alternative flows. This new research presents a new paradigm, thus making researchers, scholars, and professionals to change the use concept of BIM digital platforms. 2.- Literature review Most studies focused on the preservation of CH recognise the importance of the Information and Communication Technologies (ICTs) to generate 3D geometric models of buildings, thus constituting an essential tool for building dissemination. In addition, these technologies could 3 solve key problems when rehabilitating a heritage building, such as controlling the deformations related to the static and dynamic structure behaviour. The study of these structures using 3D geometric models is of great interest. For this purpose, techniques to acquire accurate data should be used. Most topographic methods could be used as methods to control deformations, but their use is limited to isolated cases of deflection deviation or deviation intakes. Nevertheless, the reverse engineering fully monitors the surface according to the resolution quality of the point cloud. These technologies are increasingly effective and accurate thanks to the use of computational, mathematical and experimental procedures according to both the existing scientific literature [7,8] and the technological development. The 3D terrestrial laser scanning techniques are mainly new. These techniques produce a set of digital data points in the global point cloud (GPC), where each point is represented by coordinates in the 3D space [9]. The TLS laser scanner uses the reflection of a laser beam focused on objects to calculate precisely the geometry in a short time [10]. One of the advantages of this method is obtaining a point density of the order from 5 to 10 mm, thus obtaining accurate captures of the complex geometry in the CH during the 3D geometric reconstruction. A general review of the use of this technology in the architectural heritage could complete an extensive and broad study that is focused on the goals from another perspective. In this regard, some appealing studies have used the TLS given its high level of accuracy and applicability. Some interesting research works can be obtained from Mill et al. [11]. Tang and Akinci [12] developed workflows to obtain accurate point clouds in the geometries of bridges. In the line of creating parametric objects from a point set of the TLS [13] focused the study on the tube sector and their curvature. From the point of view of the surveying engineering, laser scanner techniques have recently increased the efficiency in the creation process of parametric elements and have developed a new perspective to create BIM information models. An example of the peak being developed by this research sector is the state of the art from scan to BIM presented by Antón et al. [14], who approached to the accuracy evaluation of the 3D modelling, and Mesrop et al. [15], who created an automatic parametric workflow applied to the CH. Under this context, BIM applied to CH emerges, facing with two major challenges. On the one hand, the difficulty of having libraries containing historical architectural elements and details coinciding with architectural compositions, which are very different due to the long historical temporariness. On the other hand, the difficulty that these elements are true digital twins of the complex architectural sites. The scientific literature published until now [16–19] has been focused on implementing the grammar of forms by the interactive edition of geometry, using software supported by the visual parametric programming. These systems provide the traditional design in architecture with many functionalities. For instance, computational languages, such as Python©, carries out essential repetitive tasks (listing tables and classifying typological areas, among others) and optimises the production of the model as regards times and error reductions. The usual commands of CAD/BIM applications are reinforced by using the Application Programming Interface (API) such as Dynamo© [20], [21] and its linking to Revit Autodesk, or Grasshopper [15] in Rhinoceros, which is implemented as another tool. In addition, another BIM software such as ArchiCAD includes a direct connection with the Python language in its last version (V.24). The new applicability ArchiCAD-Python executes predefined automation scripts, and the own modification and personalisation is possible to carry out automatic tasks [22]. Another important novelty is PARAM-O [23], a visual tool that brings the designer-architect closer to 4 the Geometric Description Language (GDL) programming, thus easily creating BIM parametric objects without knowing GDL programming. However, the lack of research studies on the use of the parametric model as a data analysis tool to determine behaviours related to structural deviations caused over time justifies this paper, aiming at advancing in this line. Today the 3D modelling process is composed of several stages in an HBIM project: massive data capture with data acquisition techniques, point cloud post-processing with 2.5D captures, filtering of the point cloud to remove noise and atypical values, and segmentation of the point cloud to classify elements to be later modelled in the BIM digital platform. The next stage is the creation of parametric objects forming part of the model. Under the Revit platform, Dynamo© [24], an open graphical programming software, manages geometry parameters. ArchiCAD does the same with the direct connection and in real time with the Grasshopper programming in Rhino. Rivera et al. [21] used a point cloud and the Dynamo© as a visual programming interface to design the detail of the Royal Cross in the doorway of the Duques de Cadaval Palace in Olivenza (Badajoz). The use of these generative algorithms allows complex geometric forms to be parametrically managed, such as the parametric model of the window in the Camponeschi Palace in L´Aquila (Italy) [25], carried out in an BIM environment. Using complex geometries implies recording the characteristics of the historical architecture. In this regard, the Dynamo© script can also turn the area into a programmed parameter [26] that reflects the mapping of the surface degradation in a 3D model, semantically enriched and supported by the MySQL database [27]. The Dynamo© plugin allows the advanced treatment of the information of the model. Using programming routines based on the Python language, automatic generation actions can be carried out through point clouds [28]. This API interacting with Revit gives many advantages to parameterize the model: the programming of complex geometries, the compatibility of the mesh geometry with Revit, the creation of families, and the potential of connecting ontological knowledge [29]. The geometric modelling could be simplified to basic primitives, called as discretization processes by Bagnolo et al. [30]. This process is developed by modelling a column from the Antas Temple, dividing it into the three functional parts of the element. The methodology of building simple geometric forms with the Dynamo© script is again used in the column from the Brondo Palace (Cagliari) [31]. All these studies showed that generative algorithms allow 3D models with complex geometries to be developed. However, the analysis presented in this study (i.e., the structural analysis of a 3D model though the sequence composed of the point cloud and the parametric model) should also be considered. On the other hand, few research studies have been focused on the wooden structures of a historical building from the 16th century. The behaviour of the wooden trusses that support pitched roofs was analysed with generative algorithms to see the behaviour of the structure [32]. In addition, the Grasshopper algorithm was used, thus automatically transforming the point cloud into a 3D model, and following a workflow, both a truss model and a structural stress analysis linked to Grasshoper were performed with the functions of Dynamo-Revit. On this model representation of the complex geometry in BIM is where the difficulty of its construction lies. Various modelling strategies try that the parametric elements are digital twins. The 3D reconstruction could be generated by interpolating the point cloud into a mesh, thus creating static models with complex surfaces. Moyano et al. [33] used a flow diagram through Rhinoceros and ArchiCad to represent in an A-BIM project the orthostats of the tholos of La Pastora, an archaeological site in Seville. Moreover, Barazzetti and Previtali [34] proposed the use of neural networks through the point cloud for the geometric construction of the surface of vaults. 5 Great geometric deviations between current models and point clouds could indicate potential errors when processing models [35]. BIM information models by using MDCSs data are being more and more generated in the heritage scope. The reverse modelling process, i.e., from the point cloud to the parametric modelling, is considered in the cultural discussion in the HBIM environment [36], as well as the evaluation of the quality of heritage models in BIM, understood as the potential of informative and geometric content of 3D models [37]. On the other hand, recent studies have shown that BIM digital platforms are completely operative to study structural analyses. Nieto et al. [38] used the laser scanner both to capture architectural pieces and to analyse the deformations in a theoretical HBIM model from a structural point of view. The concept of damage implies a comparison between the current state of the structure and the state of the previous starting point [39]. The starting point could be considered as the theoretical digital model of construction without deformations, and therefore as the starting point of the analysis structure. In addition, the current state would imply the information recording of the reverse engineering process through massive data capture techniques. Structural variations could be captured by BIM operators without applying structural analysis software or other traditional research techniques or methods based on both the direct observation and simplified architectural studies. From the creation of digital twins, this research work analyses the structural deformations of a portico belonging to the courtyard of a hotel from the 18th century. TLS is used as a time-of- flight sensor, recommended for huge buildings, landscapes or archaeological studies. To avoid occlusions in the 3D survey, the points to be scanned are planned to capture as much geometry as possible. The point cloud is used as a referential auxiliary to survey the model in the BIM platform with the Revit software. Nieto et al. [40] showed that including point cloud data in the BIM software (ArchiCAD and Revit) is essential to contrast the geometric measurements and compare them with traditional techniques. To assess the quality of the model built and to analyse the structural deviations between the parametric model and the actual geometry, Dynamo© is used, and then data are converted to an .xls format. To assess the potential of the method used, structural deviations are measured with the parametric model .stl, and the point cloud in .e57 format with CloudCompare (C2C), a software for data treatment [41]. This dataset shows the capacity of both the techniques mentioned and the BIM parametric models to analyse structural deviations. The BIM methodology applied to historical buildings should consider the information variables [42], although establishing a Level of Detail (LoD) for historical buildings is not easy [25] as BIM should be related to the documentation process. The HBIM approach of the LoD is characterized by the level of detail as it is a simple or complex example within the construction domain [43]. Thus, the Levels of Development (LOD) where the agents of the process interoperate should be established in the BIM modelling process. LOD is a term referring to the amount of information of the elements that are part of BIM. According to the uBIM guide, LOD levels systematize and unify the degree of reliability of the information in the HBIM model [44]. In this regard, this study has been focused on a LOD 250 approach by defining various surface construction levels, such as cornices and decorative elements, without addressing the internal constructive part of the elements. 3.- Methodology 3.1.- The case study of the Arizón Palace 6 In the 15th century, the area of the margin of the coast of Cádiz, together with Jerez de la Frontera, was economically developed by the American Conquest. The city of Sanlúcar de Barrameda (Cádiz), located in the mouth of the Guadalquivir river, increased demographically, increasing outside of the primitive wall. The area was increased by the singularity of the maritime trade as the merchants used the harbour in Sanlúcar de Barrameda to start their journeys to the West Indies. The trade development implied an opportunity for rich people as they related to the trade traffic with the American colonies. This development was embellished not just in the urban development but in the construction of architectural sites of historical value. The complexity of the architectural site of the Arizón Palace from the 17th century is only understood by the perception of the trade traffic with America [45]. This architectural site was the house belonging to merchants to the Indies and used for storage; the nobles also lived there. The Courtyard of Honour (see Figure 1) was built between 1721 and 1727. Given the increasing beach tourism, the palace was fully rehabilitated in 1992 and declared of Cultural Interest in 2001. Figure 1. View of the courtyard of the Arizón Palace: a) Photograph of the Courtyard of Honour in 1989 and detail of the central well; and b) Photograph of the Courtyard of Honour in 2009 at the beginning of the last rehabilitation. 3. 2.- Data Collection Massive data of the space were acquired with various topographic techniques. The methods based on both images and the 3D laser scanning are the most used in large projects as the old traditional methods, manual methods such as the tape measure or the laser measure, provide many errors. Today the data acquisition technique based on TLS is the most used as it provides accuracy and speed, although other methods based on images have also been used to rebuild 3D models because of their economic advantages, effectiveness [46], and accuracy when modelling CH in BIM [47]. To detect and determine the distance with laser images (LIDAR), TLS differs from operating in an aerial platform. Its methodology is based on calculating the distance between the laser and the object. This procedure is developed by using the time-of- flight method or the wave transmitted and received by the signal [48]. The method scans all the surface area, capturing thousands of points in a coordinate system (x, y, z) and obtaining the point cloud. To know the influence of the geometric variables, topographic tools should be used to measure the accuracy of the solids to be represented. The validity of the model is no more based on the establishment of 3D coordinates (x,y,z) of the GCP, thus guaranteeing that the model is correctly scaled [47]. Together with the laser scanner, the Leica Flexline TS02 total station was used with an accuracy of 2 mm [49] to record the coordinates of the GCP in the 7 four façades plans of the courtyard. Although the area of the courtyard of the building is not a complex structure, unlike a bridge where the topographic geometry must be deeply studied, the GPC recorded was referenced by points of the total station that provides the other points of the station with a UTM-29S coordinate basis in the WGS 84 system. Given its compositional wealth, the Courtyard of Honour with porticos in the Arizón Palace was chosen for both the study and the metric-architectural survey (see Figure 2a). A Leica BLK360 laser scanner [50] was used to capture all the geometry of the space. Considering, therefore, that the courtyard provided appropriate dimensions for the scope of 60 range metres, this scanner was used because of its light weight and convenience to be carried. The transversal length of the site analysed for the modelling is not greater than 8.89 metres. In other studies, [51] [52] and [53] used this equipment for greater distances, and [54] showed that the manufacture specifications are averagely fulfilled. The Leica BLK360 laser scanner uses Waveform Digitising technology (WFD), with a maximum scanning speed of 360,000 points/second. It has three HDR digital cameras of colour sensor and fixed focal length (simple image 2592 x 1944 pixels, 60° x 45° (v x hz), full-dome scan of 30 images and automatic rectification of space, 150 Mpx, 360° x 300°), as well as an infrared thermal camera (simple image 160 x 120 pixels, 71° x 56° (v x hz), full-dome scan of 10 images, 360° x 70°), the four included in the equipment. The laser system of the equipment obtains a range accuracy from 4 mm to 10 m, and from 7 mm to 20 m. Figure 2. 3D scanning process: a) Data acquisition for control points; and b) Point Cloud Layout with the equipment position. The geometry captured by the laser scanner was previously treated, reconsidering the optimal equipment position, so a full scanning could record the spaces of the four sides of the courtyard and its galleries. The itinerary took place regularly to reduce the hidden geometry and occluded points [55] , thus obtaining a point density as homogeneous as possible (see Figure 2b). The TLS scanning process was carried out through five placements in the strategic points of the four galleries: corners and access through the stairs, and other three in the courtyard surrounding the central well. The distribution of the equipment implied a distance between the laser scanner and the object from 4 to 7 metres (Figure 3). The “High density” mode was used in all the scans, with a full domo scan duration of 3:40 min, summing 2:30 min 8 from capturing HDR images. The equipment performance was studied by [56], which confirmed an accuracy of the scanner from 6 mm to 10 metres to record control points. Figure 3. Section/perspective of the Point Cloud with the distribution of the 8 equipment and the distance among them. Visualisation from Leica JetStream Viewer. After obtaining the records, the adjustment post-processing and cleaning were carried out by the Leica Register 360 software, creating the GPC. Both outside the courtyard and inside galleries, a coupling and alignment of the point cloud were used through effective and accurate procedures, thus keeping an overlay among all positions. A total of eight surveys were conducted at one level, thus fulfilling the goal of covering most of the space of the Courtyard of Honour. The itinerary took place as regular as possible to reduce errors and achieve an optimal point density [57]. A total of 15 links were enough to ensure the scanning coupling (¡Error! No se encuentra el origen de la referencia.). Table 1.- Result of the records with error expression in the placement set. The recording was carried out through the automatic alignment of the Leica Cyclone REGISTER 360 software, whose alignment resulted in 43 scanning connections among the various point clouds. Finally, a point cloud of 103 million points was obtained for a file size of 10.6 GB in a non-structured format (.e57). Data were also exported to the .pts format of 15.05 GB. 4.- Research workflow 4.1.- Introduction The action protocol of this research requires clear goals before modelling the heritage assets. Techniques to acquire massive data, such as TLS, were used to fully control the geometry of the objects and the furniture of the historical buildings. The point cloud provides the characteristics and the geometric and chromatic information of the materials [58], the TLS placements Number: Number of connections: Strength: Overlapping: Set error: Cloud to cloud error: Level 1 Survey 1 15 placements 43 64% 51% 0.004 m 0.004 m 15 This visual programming tool can visually guide the behaviour and define the geometries through commands for BIM operators. This action takes place in real-time and, with few programming knowledge, a dataset could be obtained by analysing the structural deformations. In addition, as developed in the programming nodes, a colour-scale map could be included, visually determining the deviations between the building model and the point cloud. To validate the research model in the effectiveness of the study developed with the BIM digital platform through Revit with Dynamo©, the model and the point cloud was recorded within a reference framework through the algorithm of C2C©. The experimental sample included three architectural elements built in the general model of the Courtyard of Honour in the Arizón Palace. Thus, the independent column C1 could be studied. In this column, the geometric quality of the construction of the shaft and its respective morphological elements of base and capital can be determined. In addition, the column and a part of the arch could also be studied to analyse the structural deviations between these two elements. Their comparison could determine whether the quality of the built model is adjusted to reality, as well as the fundamental variations. Finally, the northern portico was fully studied, establishing the total deviation and levelling values of the model. C2C© has been widely used to compare the point cloud obtained with SfM, taking TLS as a model: Georgantas et al. [77] used TLS as a reference to compare the IBM photogrammetry through C2C©; Teza et al. [78] compared TLS and SfM to determine a morphological analysis in terms of differences. C2C© was also used to analyse the image reconstruction algorithms obtained with Agisoft and Open Computer Vision (CV) [79]; Koutsoudis et al. [80] analysed the average distances and the standard deviations of all the single-view range scans in comparison with the mesh; and Moyano et al. [33] compared the construction of BIM parametric objects belonging to complex archaeological forms with the point cloud data from the TLS to assess the modelling. In addition, Robles et al. [81] analysed the verticality of the Bayacas bridge in Órgiva using CloudCompare® software to compare the point cloud with a vertical plane. A similar method was experimented by Nieto et al. [82], who evaluated the deformation of a roof beam of the Pavilion of Charles V by overlapping the point cloud and the HBIM. Used as a reference, the point cloud was figuratively a photograph of the damaged area. Moreover, 0 200 400 600 800 1000 1200 1400 Number of points Error (metres) Column E1 C2M signed distances (14,340 values) 16 textual markers (‘Marker’ tool in ArchiCAD®) were incorporated alongside parametric objects to indicate structural elements affected by water seeping. Andriasyan et al. [15] analysed the study between the model and the point cloud of Scan to BIM in the courtyard of the Pilatos Palace in Seville. Makupa et al. [83] showed both the importance of the TLS to detect structural changes and to monitor deformations and how the M3C2C algorithm developed by [84] should be used as a comparative method for deformations. To compare the building model in BIM with the point cloud of the TLS, the total point cloud was worked with 107 million points in a non-structured format (.pts) and including it in C2C© to segment it in three parts: column C1, set of column and arch, and northern portico. After the segmentation, noise was filtered and cleaned according to the parameters established by Girardeau [41], resulting in three point subsets with a structured-binary format (.e57). The segmentation was manually carried out, and part of the data subset segmented with the HBIM model was included in the software in the .stl format Polygon fencing in software CloudCompare© Version 2.10-alpha [85], which removed noise, non-related points and nondesirable elements, such as seats placed in the middle of the courtyard, floor lamps, and flowerpots. After obtaining the point subsets of the point cloud of the segmented TLS, the HBIM building model was aligned with each part (see Figure 10) through the pairs of points selected (see Figure 11) which are common between the .stl model and the .e57 point cloud. The six pairs of points (northern portico) (R0-A0, R1-A1, R2-A2, R3-A3, R4-A4, and R5-A5) were located in the capitals and the base of the intermediate columns, with errors of 0.015, 0.017, 0.016, 0.015, 0.011, and 0.014 respectively, and an RMS of 0.01542 m. The co-recording or recording was applied with the Iterative Closest Point (ICP) algorithm, implemented by CloudCompare M3C2 and created by Lague et al. [84] to optimise the alignment automatically. The results of the series of the three point subsets are included in Table 3. Figure 11. Alignment of the point cloud of the northern portico, and the 3D model extracted by Revit. 17 Table 3.- Results of the RMS of the alignment between the parametric model and the point cloud. RMS of the alignment between the parametric model and the point cloud in C2C© Column C1 Set of column and arch Portico RMS 0.0076 0.0061 0.0154 Thus, the distance between the model and the subset of partial points of the segmented portico was calculated. The parameters studied were the Root Mean Square Error (RMSE) according to Equation 1, the minimum and maximum distances between the model and the point cloud, the average distance, the standard deviation, and the estimated error in metres. The deviation between the digital twin model and the actual geometry presented two characteristics: the high presence of points in the zero value in comparison with the other distance intervals, and the high standard deviation that could be calculated through the equation already mentioned by Arias et al. [86]: E.1 𝑅𝑀𝑆𝐸= √∑(∆𝑋𝑖)2+(∆𝑌𝑖)2+(∆𝑍𝑖)2 𝑛−1 𝑛 𝑖=1 Where: o ΔX; ΔY; ΔZ are the differences between the coordinates of the point cloud and the coordinates of the closest points of the 3D model. o “N” is the number of control points. To validate the programming of the deviations obtained with Dynamo©, the frequency of atypical Gauss values was also analysed. This frequency showed the statistical approach of the behaviour of the point distribution according to Equation 2. E.2 𝑌= 1 𝜎√2𝜋𝑒−(𝑥−𝜇)² 2𝜎² Where: - σ = standard deviation - σ² = variance - µ = mean - π = Pi number - e = Euler number - Y = height of the curve for any value of x - x = value in the horizontal axis The frequency data distribution with respect to the height and the deviations in metres were estimated with a Gaussian function density, with a bandwidth of 10. 18 7.- Results obtained in the geometric assessment 7.1.- Results from the interpretation of the deviations with Dynamo© Under the dataset provided by Dynamo©, the parameters studies were those correlated to the direct data emitted by the use basis of the M3C2C algorithm. These parameters were as follows: the minimum distance, the maximum distance, the average distance, and the standard deviation. After exporting data from Dynamo© to a structured format (.xls), data were treated to obtain the parameters according to the results included in Table 4. Table 4.- Average distances and standard deviations of the single-view range scans compared between the building model and the point cloud (in metres). Average distances and standard deviations of the single-view range scans compared between the building model and the point cloud (in metres) Survey A (Dynamo©) Standard deviation Min. Distance Max. Distance Average distance Column C1 0.0074 0.0000 0.0396 -0.0033 Set of column and arch 0.0407 0.0000 0.1906 0.0265 Portico 0.0446 0.0000 0.5520 0.0109 7.1.1- Analysis of the column For the working process using the code of the nodal algorithm created, after included in the Dynamo environment, the execution of the algorithm was changed from automatic to manual to avoid keeping an unfeasible code in an automatic execution. The “Select Model Element” node was used to select the column to be studied, the prismatic mass previously created, and the point cloud. Through the Element.Faces, the “column” element was turned into a geometry interpreted by Dynamo©, and then an axis point (PE) of the resulting geometry was obtained. On the other hand, the point cloud was segmented by using the prismatic mass and decimated by the user. The resulting set of points was called Np. This process implied that the selection was like the segmentation of a point subset within the decimated point cloud. Coordinate Z was removed from PE, maintaining the coordinates X and Y. Coordinates Z were extracted from Np and coupled with PE (x, y), thus obtaining points both in the axis of the geometry and in the point cloud. These points created on the axis of the geometry were called Ez”. The vector 𝐸𝑍𝑁𝑃 → was created, and following the direction of this vector, the points of Np were transferred to a distance that ensured that none of these points was within the geometry (Np’). The vector 𝑁𝑃𝐸𝑍 → was created, and following its direction, the points of Np’ were projected on the geometry. Through this process, the nearest point of the 3D model to its respective neighbour of the point cloud was determined. This point set on the 3D model was called Pm. A line with the points of Np and Pm was created, and their length were extracted, which coincided with the absolute minimum distances between the point cloud and the 3D model. To determine the negative distances, or in other words, the points of Np within the model, the vector 𝑁𝑃𝑃𝑚 → was created, and the mathematical condition was imposed according to Equation 3: 19 E.3 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒={+, 𝑠𝑖 |𝐸𝑍𝑁𝑃 → |>|𝑁𝑃𝑃𝑚 → +𝐸𝑍𝑁𝑃 → | −, 𝑠𝑖 |𝐸𝑍𝑁𝑃 → |<|𝑁𝑃𝑃𝑚 → +𝐸𝑍𝑁𝑃 → | Figure 12 graphically represents the vectors to which the mathematical condition of Equation 3 refers, and Figure 13 shows the code of the nodal algorithm executed on the 3D model of the column C1 of the portico. The colour gradient was applied to the lines obtained in the nodal process. Figure 12. Graphic representation of the code process of the nodal algorithm. Figure 13. Parametric modelling of the central column of the northern portico. 20 Table 5.- Histogram of the column C1 (Dynamo©). Table 6.- Graph of the normal deviation of the column C1 (Dynamo©). The column image showed that there were certain residual values (see Figure 13) corresponding to the point cloud being part of the flowerpots, and the graph included in Table 2 shows that approximately 80% of the sample points were between -0.01 and 0.01 m. Thus, there was no systematic influence within the algorithm, and the deviations were not significant. Table 6 shows a deviation tendency of -0,004 m in the point cloud in comparison with the 3D model. In the analysis, the negative values corresponded to those points within the model, and the positive were outside the model in both sides. 7.1.2.- Analysis of the set of column and arch. 0 500 1000 1500 2000 2500 3000 3500 Number of points Metres Column C1 C2M signed distances (13,745 values) 0 10 20 30 40 50 60 -0,03 -0,02 -0,01 0 0,01 0,02 0,03 0,04 0,05 Frequency Metres NORMAL 21 a) b) Figure 14. a) Parametric modelling of the column C1, and b) beginning of the arch of the column C1. In this case there were also atypical or residual values in the analysis, such as the case of the values corresponding to the drainpipe found in the segmentation of the arches. The reason was that the point cloud could not be accurately segmented within Dynamo©. Table 7.- Histogram of the set of column and arch. Unlike the previous case, Table 7 shows the existence of significant deviations from the value of 0.020 m. These values were the result of the external elements of the assessment model, and therefore they were residual values. As Figure 14b shows, the points with the greatest deviation in comparison with the model studied corresponded to the drainpipe, and the proportionality of these residual values in comparison with the values corresponding to the model studied was greater than in the previous case. Moreover, this increase from the value of 0.055 m took place because the ornaments of the wall were not considered in the 3D model; the flat sides of the wall were only considered. In 0 1000 2000 3000 4000 5000 6000 -0,02 -0,01 0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09 0,1 0,11 0,12 0,13 0,14 0,15 0,16 0,17 0,18 0,19 Number of points Metres Column-arch C2M signed distances (38,745 values) 22 the graphic result of the normal deviation (Table 7), the average of the deviation was virtually 0. Table 8.- Graph of the normal deviation of the set of column and arch (Dynamo©). 7.1.3.- Analysis of the column portico Figure 15 Parametric modelling of the northern portico. a) b) c) d) e) 0 2 4 6 8 10 12 -0,05 0 0,05 0,1 0,15 0,2 0,25 Frequency Metres NORMAL 23 Figure 16. a) Column A1, b) Column B1, c) Column C1, d) Column D1, and e) Column E1. As expected, this case also obtained residual values; however, the segmentation was greater, so the proportionality of these values was not so significant as in the previous case. Table 9.- Histogram of the northern portico of the Courtyard of Honour. There was also an increase from the value of 0.055 m, but it was lower than in the case of Table 7. In addition, almost all points were between -0.01 and 0.02 m. 0 5000 10000 15000 20000 Number of points Metres Row 1 portico C2M signed distances (98,955 values) 24 Table 10.- Graph of the normal deviation of the northern portico (Dynamo©). 7.2.- Results from the interpretation of the deviations with C2C© Under the dataset provided by C2C©, the parameters studies were those correlated to the direct data emitted by the use basis of the M3C2C algorithm. These parameters were as follows: the minimum distance, the maximum distance, the average distance, and the standard deviation. The difficulty presented by the point cloud post-processing in both BIM and another platform was the file extension. For instance, ArchiCAD® Graphisoft presented a limit of 14 MB, which is overcome in Revit® Autodesk by the workflow with ReCap. In this case, the cloud was processed though the Cyclone Register 360 software and was exported to an interchange format (.pts) to maintain the filtering and the segmentation applied. Cloud Compare is a software for point cloud processing and is capable of segmenting and filtering, but the size of the .pts file (14 GB) slows down the work. From the respective segmentation of the point subsets, they were kept in non-structured files to be compared between the 3D model and the point subsets being part of the column C1, the set of column and arch, and the portico. Table 11 shows the data obtained by following the parameters described. Table 11.- Average distances and standard deviations of the single-view range scans compared between the building model and the point cloud (in metres). Average distances and standard deviations of the single-view range scans compared between the building model and the point cloud (in metres) Survey B (C2C©) Standard deviation Min. Distance Max. Distance Average distance RMS Estimated standard error Column C1 0.0092 0.0000 0.0668 0.0087 0.0076 0.0092 Set of column and arch 0.0185 0.0000 0.1146 0.0115 0.0061 0.0204 Portico 0.0201 0.0000 0.1608 0.0540 0.0154 0.0402 0 1 2 3 4 5 6 7 8 9 10 -0,1 0 0,1 0,2 0,3 0,4 0,5 0,6 Frequency Metres NORMAL 31 reconstructions. By creating digital twins, this research work analyses the structural deformations in a portico belonging to the courtyard of a hotel from the 18th century. TLS, recommended for large buildings, was used as a technique to acquire accurate data. Moreover, the point cloud was used as a referential auxiliary to build the model in the BIM platform with the Revit software. To assess the quality of the model built and to analyse the structural deviations between the parametric model and the actual geometry, Dynamo© was used. To validate the experimentation, structural deviations were measured with both the parametric model (.stl) and the point cloud in the .e57 format with CloudCompare, a software for data treatment. Based on this dataset, the deviation of the results obtained with Dynamo© and C2C© in the quartiles Q1 and Q3 ranged between 0.5 and 1.7 cm in the most unfavourable case (in the column C1). Compared with the length of the elements, the dispersion could be acceptable, so the techniques mentioned could be used to review the visual recordings and to analyse structural deviations when building parametric objects. These analysis records could take place in various periods, and the displacement analysis could be seen through an action protocol. Movements recorded deviations in the walls between 2 and 5 cm, which can be considered usual in old buildings because of their characteristics and their historical identity. In addition, the comparison between the model of the point cloud and the parametric model allowed the quality of the model built to be known; effectiveness should also be improved, and the ornamental elements not detected today by the algorithm should be listed. Dynamo© is a visual programming application used to solve many tasks, including the analysis and the creation of routines related to the geometric layout. Thus, data acquisition could be transferred to programs such as Excel© to be treated or parametrically designed, becoming them versatile for BIM operators. Assessing the dispersions among homologous plans in the same platform is among the advantages of this algorithm, and in other studies the linear analysis was assessed through an independent software. A digital twin reproduces the behaviour of an actual system as response to external physical data [32]. Thus, this algorithm could be a useful tool for both programmers and engineers in relation to structures modelled with BIM and architects who conduct structural analysis campaigns in old buildings. This case study shows that the proposed methodology is greater effective to analyse a structural model. It is both an analysis model as it contains geometric information about the position, size and space to know the mechanical properties, and a detail model where the geometric forms of the structural elements are included [89]. A future research line will be focused on using the SfM technique in the analysis, behaviour, and diagnosis of humidities, as well as in the experimentation related to the distortion, axils and bending of architectural elements. Given the complex scope of study that CH involves, future studies are expected to focus on more complex architectural structures and elements to stress the value of the research. Thus, the new BIM paradigm would be included in an interdisciplinary programming within the area of architectural structures. Abbreviations The following abbreviations are used in this manuscript: TLS Terrestrial Laser Scanning SfM Structure-from-Motion MDCSs Massive Data Capture Systems 32 BIM Building Information Modelling HBIM Historic Building Information Modelling GCP Ground Control Points DBAT Damped Bundle Adjustment Toolbox RPAS Remotely Piloted Aircraft Systems DSLR Digital Single Lens Reflex HDR High Dynamic Range RMS Root-Mean-Square RMSE Root-Mean-Square-Error MVS Multi-view Stereo EXIF Exchangeable image file format References [1] X. Xu, J. Bureick, H. Yang, I. Neumann, TLS-based composite structure deformation analysis validated with laser tracker, Compos. Struct. 202 (2018) 60–65. https://doi.org/10.1016/j.compstruct.2017.10.015. [2] R. Jiang, D. V. Jáuregui, K.R. White, Close-range photogrammetry applications in bridge measurement: Literature review, Meas. J. Int. Meas. Confed. 41 (2008) 823–834. https://doi.org/10.1016/j.measurement.2007.12.005. [3] M. De Fino, R.A. Galantucci, F. Fatiguso, Remote diagnosis and control of heritage architecture by photorealistic digital environments and models, SCIRES-IT. 9 (2019) 1– 16. https://doi.org/10.2423/i22394303v9n2p1. [4] F. Banfi, HBIM, 3d drawing and virtual reality for archaeological sites and ancient ruins hbim, dibujo 3d y realidad virtual aplicados a sitios arqueológicos y ruinas antiguas 3d drawing and virtual reality for archaeological sites and ancient ruins., Virtual Archaeol. Rev. 11 (2020) 16–33. https://doi.org/10.4995/var.2020.12416. [5] C. Chevrier, N. Charbonneau, Parametric documenting of built heritage: 3D virtual reconstruction of architectural details, International. (2010). http://multiscience.atypon.com/doi/abs/10.1260/1478-0771.8.2.135 (accessed May 10, 2016). [6] M. Murphy, E. McGovern, S. Pavia, Historic Building Information Modelling - Adding intelligence to laser and image based surveys of European classical architecture, ISPRS J. Photogramm. Remote Sens. 76 (2013) 89–102. https://doi.org/10.1016/j.isprsjprs.2012.11.006. [7] J. Wang, W. Sun, W. Shou, X. Wang, C. Wu, H.Y. Chong, Y. Liu, C. Sun, Integrating BIM and LiDAR for Real-Time Construction Quality Control, J. Intell. Robot. Syst. Theory Appl. 79 (2015) 417–432. https://doi.org/10.1007/s10846-014-0116-8. [8] T. Medić, H. Kuhlmann, C. Holst, Designing and evaluating a user-oriented calibration field for the target-based self-calibration of panoramic terrestrial laser scanners, Remote Sens. 12 (2020) 15. https://doi.org/10.3390/RS12010015. [9] S. Slob, R. Hack, 3D Terrestrial Laser Scanning as a New Field Measurement and Monitoring Technique, in: Springer, Berlin, Heidelberg, 2004: pp. 179–189. https://doi.org/10.1007/978-3-540-39918-6_22. [10] V. Gikas, Three-Dimensional Laser Scanning for Geometry Documentation and Construction Management of Highway Tunnels during Excavation, Sensors. 12 (2012) 33 11249–11270. https://doi.org/10.3390/s120811249. [11] T. Mill, A. Alt, R. Liias, Combined 3D building surveying techniques-Terrestrial laser scanning (TLS) and total station surveying for BIM data management purposes, J. Civ. Eng. Manag. 19 (2013). https://doi.org/10.3846/13923730.2013.795187. [12] P. Tang, B. Akinci, Formalization of workflows for extracting bridge surveying goals from laser-scanned data, in: Autom. Constr., Elsevier, 2012: pp. 306–319. https://doi.org/10.1016/j.autcon.2011.09.006. [13] U. Bauer, K. Polthier, Generating parametric models of tubes from laser scans, CAD Comput. Aided Des. 41 (2009) 719–729. https://doi.org/10.1016/j.cad.2009.01.002. [14] D. Antón, B. Medjdoub, R. Shrahily, J. Moyano, Accuracy evaluation of the semiautomatic 3D modeling for historical building information models, Int. J. Archit. Herit. 12 (2018) 790–805. https://doi.org/10.1080/15583058.2017.1415391. [15] M. Andriasyan, J. Moyano, J.E. Nieto-Julián, D. Antón, From Point Cloud Data to Building Information Modelling: An Automatic Parametric Workflow for Heritage, Remote Sens. 2020, Vol. 12, Page 1094. 12 (2020) 1094. https://doi.org/10.3390/RS12071094. [16] E. Touloupaki, T. Theodosiou, Energy Performance Optimization as a Generative Design Tool for Nearly Zero Energy Buildings, in: Procedia Eng., Elsevier Ltd, 2017: pp. 1178–1185. https://doi.org/10.1016/j.proeng.2017.04.278. [17] C. Sydora, E. Stroulia, Rule-based compliance checking and generative design for building interiors using BIM, Autom. Constr. 120 (2020) 103368. https://doi.org/10.1016/j.autcon.2020.103368. [18] B. Ferreira, A. Leitão, Generative Design for Building Information Modeling, BIM-Early Des. 1 (2015) 635–643. [19] W. Wahbeh, Building skins, parametric design tools and BIM platforms, in: Conf. Proc. 12th Conf. Adv. Build. Ski., 2017: pp. 1104–1111. https://www.researchgate.net/publication/320244444 (accessed February 4, 2021). [20] M. Capone, E. Lanzara, Scan-to-BIM vs 3D ideal model HBIM: Parametric tools to study domes geometry., ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 42 (2019) 219–226. [21] B. Rivera, P. Merchán, S. Salamanca, E. Pérez, M.D. Moreno, M.J. Merchán, Creación de bibliotecas de objetos paramétricos para su integración en modelo HBIM, in: Actas Las XXXIX Jornadas Automática, Badajoz, 5-7 Septiembre 2018, 2018: pp. 1069–1076. [22] Graphisoft, Python – GRAPHISOFT, (2019). https://graphisoft.com/downloads/python (accessed February 1, 2021). [23] Graphisoft, PARAM-O. Creación de objetos paramétricos para ArchiCAD, (2021). https://graphisoft.com/es/downloads/param-o (accessed February 1, 2021). [24] Autodesk, Dynamo BIM, (2016). https://dynamobim.org/ (accessed March 4, 2020). [25] S. Brusaporci, P. Maiezza, A. Tata, Computational Design for As-Built Modeling of Architectural Heritage in HBIM processes, in: 2018 IEEE Int. Conf. Metrol. Archaeol. Cult. Heritage, MetroArchaeo 2018 - Proc., Institute of Electrical and Electronics Engineers Inc., 2018: pp. 199–203. https://doi.org/10.1109/MetroArchaeo43810.2018.13620. [26] C. Santagati, C.R. Laurini, G. Sanfilippo, N. Bakirtzis, D. Papacharalambous, S. Hermon, Hbim for the surveying, analysis and restoration of the saint john the theologian cathedral in Nicosia (Cyprus), ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 42 (2019) 1039–1046. https://doi.org/10.5194/isprs-Archives-XLII-2-W11-1039-2019. [27] R. Quattrini, C. Battini, R. Mammoli, HBIM to Vr. Semantic awareness and data enrichment interoperability for parametric libraries of historical architecture, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. - ISPRS Arch. 42 (2018) 937–943. 34 https://doi.org/10.5194/isprs-archives-XLII-2-937-2018. [28] R. Angulo-Fornos, M. Castellano-Román, HBIM as Support of Preventive Conservation Actions in Heritage Architecture. Experience of the Renaissance Quadrant Façade of the Cathedral of Seville, Appl. Sci. 10 (2020) 2428. https://doi.org/10.3390/app10072428. [29] X. Yang, M. Koehl, P. Grussenmeyer, Mesh-to-bim: From segmented mesh elements to bim model with limited parameters, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. - ISPRS Arch. 42 (2018) 1213–1218. https://doi.org/10.5194/isprs-archives-XLII-2-1213- 2018. [30] V. Bagnolo, R. Argiolas, A. Cuccu, HBIM for archaeological sites: from SFM based survey to algorithmic modeling, Int. Arch. Photogramm, Remote Sens. Spat. Inf. Sci. XLII–2 (2019) 57–63. https://doi.org/10.5194/isprs-archives-XLII-2-W9-57-2019. [31] V. Bagnolo, R. Argiolas, A. Cuccu, Digital survey and algorithmic modeling in HBIM. Towards a library of complex construction elements., Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. - ISPRS Arch. 42 (2019) 25–31. https://doi.org/10.5194/isprsarchives-XLII-4-W12-25-2019. [32] A. Massafra, D. Prati, G. Predari, R. Gulli, Wooden Truss Analysis, Preservation Strategies, and Digital Documentation through Parametric 3D Modeling and HBIM Workflow, Sustainability. 12 (2020) 4975. https://doi.org/10.3390/su12124975. [33] J. Moyano, C.P. Odriozola, J.E. Nieto-Julián, J.M. Vargas, J.A. Barrera, J. León, Bringing BIM to archaeological heritage: Interdisciplinary method/strategy and accuracy applied to a megalithic monument of the Copper Age, J. Cult. Herit. (2020). https://doi.org/10.1016/j.culher.2020.03.010. [34] L. Barazzetti, M. Previtali, Vault modeling with neural networks, ISPRS Ann. Photogramm. Remote Sens. Spat. Inf. Sci. 42 (2019) 81–86. https://doi.org/10.5194/isprs-archives-XLII-2-W9-81-2019. [35] E.B. Anil, P. Tang, B. Akinci, D. Huber, Deviation analysis method for the assessment of the quality of the as-is Building Information Models generated from point cloud data, Autom. Constr. 35 (2013) 507–516. https://doi.org/10.1016/j.autcon.2013.06.003. [36] L. Paris, W. Wahbeh, Survey and representation of the parametric geometries in HBIM, DISEGNARECON. 9 (2016) 12.1-12.9. http://disegnarecon.univaq.it/ojs/index.php/disegnarecon/article/view/164 (accessed January 14, 2021). [37] A. Fryskowska, J. Stachelek, A no-reference method of geometric content quality analysis of 3D models generated from laser scanning point clouds for hBIM, J. Cult. Herit. 34 (2018) 95–108. https://doi.org/10.1016/J.CULHER.2018.04.003. [38] J.E. Nieto-Julián, D. Antón, J.J. Moyano, Implementation and Management of Structural Deformations into Historic Building Information Models, Int. J. Archit. Herit. 0 (2019) 1– 14. https://doi.org/10.1080/15583058.2019.1610523. [39] J.P. Martínez, H.A. Rivera, J.Á.O. Lozano, J. Luis, G. Ruvalcaba, M.Z. de León, T.C. Hernández, M. Navarro, Hernández, Gerardo Arayza Garaygordobil, Las vibraciones como una fuente para evaluar el comportamiento estructural de edificios patrimoniales, Soc. Mex. Ing. (2012) 1–18. http://www.redalyc.org/articulo.oa?id=61851598002. [40] J.E. Nieto, J.J. Moyano, F. Rico, D. Antón, Management of built heritage via HBIM Project: A case of study of flooring and wall tiling, Virtual Archaeol. Rev. 7 (2016) 1–12. https://doi.org/10.4995/var.2016.4349. [41] D. Girardeau-Montaut, CloudCompare Point Cloud Processing Workshop, (2019) 1–38. www.cloudcompare.org@CloudCompareGPL (accessed January 28, 2021). [42] P. Parisi, M. Lo Turco, E.C. Giovannini, The value of knowledge through H-BIM models: historic documentation with a semantic approach, (2019). https://doi.org/10.5194/isprsarchives-XLII-2-W9-581-2019. 35 [43] R. Brumana, S. Della Torre, M. Previtali, L. Barazzetti, L. Cantini, D. Oreni, F. Banfi, Generative HBIM modelling to embody complexity (LOD, LOG, LOA, LOI): surveying, preservation, site intervention—the Basilica di Collemaggio (L’Aquila), Appl. Geomatics. 10 (2018) 545–567. https://doi.org/10.1007/s12518-018-0233-3. [44] E. Nieto, J.J. Moyano, Á. García, Estudio constructivo del Palacio de los Niños de Don Gome (Andújar, Jaén), gestionado desde el proyecto HBIM, Virtual Archaeol. Rev. 10 (2018) 84. https://doi.org/10.4995/var.2019.10567. [45] F. Olaciregui, Rehabilitacin del conjunto arquitectnico del Marqus de Casa Arizn (S. XVII-S. XVIII) en Sanlcar de Barrameda, Cdiz : leccin de ingreso como Amigo de Nmero leda el da 25, Delegacin en Corte Departamento de Publicaciones, Madrid, 2015. [46] Q. Lu, S. Lee, Image-Based Technologies for Constructing As-Is Building Information Models for Existing Buildings, J. Comput. Civ. Eng. 31 (2017) 04017005. https://doi.org/10.1061/(asce)cp.1943-5487.0000652. [47] J. Moyano, J.E. Nieto-Julián, D. Bienvenido-Huertas, D. Marín-García, Validation of Close-Range Photogrammetry for Architectural and Archaeological Heritage: Analysis of Point Density and 3d Mesh Geometry, Remote Sens. 12 (2020) 3571. https://doi.org/10.3390/rs12213571. [48] H.M. Yilmaz, M. Yakar, S.A. Gulec, O.N. Dulgerler, Importance of digital close-range photogrammetry in documentation of cultural heritage, J. Cult. Herit. 8 (2007) 428–433. https://doi.org/10.1016/j.culher.2007.07.004. [49] N. Yastikli, Documentation of cultural heritage using digital photogrammetry and laser scanning, J. Cult. Herit. 8 (2007) 423–427. https://doi.org/10.1016/j.culher.2007.06.003. [50] L. Geosystems, Leica Cyclone REGISTER 360-3D Laser Scanning Point Cloud Registration Software, (2018). https://doi.org/https://leicageosystems.com/products/laser-scanners/software/leica-cyclone/leica-cyclone-register- 360. [51] C. León-Robles, J. Reinoso-Gordo, J. González-Quiñones, Heritage Building Information Modeling (H-BIM) Applied to A Stone Bridge, ISPRS Int. J. Geo-Information. 8 (2019) 121. https://doi.org/10.3390/ijgi8030121. [52] D. Ronchi, … M.L.-I.A. of, undefined 2019, FIELD WORK MONITORING AND HERITAGE DOCUMENTATION FOR THE CONSERVATION PROJECT. THE" FORO EMILIANO" IN TERRACINA (ITALY)., Search.Ebscohost.Com. (n.d.). http://search.ebscohost.com/login.aspx?direct=true&profile=ehost&scope=site&authtype =crawler&jrnl=16821750&AN=139412089&h=KAcejLCd1QhSFE5VlM8LVkTCFLW0l80h heJd41ZHQNLEt2IaeuHT%2FpRKsKGqSgL9NKsfvxe4YjdBUhDqqHQNyQ%3D%3D&cr l=c (accessed March 29, 2021). [53] T. Luhmann, M. Chizhova, D. Gorkovchuk, H. Hastedt, N. Chachava, N. Lekveishvili, COMBINATION OF TERRESTRIAL LASERSCANNING, UAV AND CLOSE-RANGE PHOTOGRAMMETRY FOR 3D RECONSTRUCTION OF COMPLEX CHURCHES IN GEORGIA, (2019). https://doi.org/10.5194/isprs-archives-XLII-2-W11-753-2019. [54] M. Lindstaedt, T.P. Kersten, R. Blaskow, D. Schneider, T. Kersten, Untersuchungen zum Genauigkeitspotential des terrestrischen Laserscanners Leica BLK360 Comparative Geometrical Investigations of Hand-held Scanning Systems (Update) View project Virtual Reality View project Untersuchungen zum Genauigkeitspotential des terrestrischen Laserscanners Leica BLK360, 2018. https://www.researchgate.net/publication/325346976 (accessed March 29, 2021). [55] M. Bonduel, M. Bassier, M. Vergauwen, P. Pauwels, R. Klein, Scan-to-bim output validation: Towards a standardized geometric quality assessment of building information models based on point clouds, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. - ISPRS Arch. 42 (2017) 45–52. https://doi.org/10.5194/isprs-archives-XLII-2-W8-45-2017. 36 [56] M. Lindstaedt, T.P. Kersten, R. Blaskow, D. Schneider, T. Kersten, Untersuchungen zum Genauigkeitspotential des terrestrischen Laserscanners Leica BLK360 Comparative Geometrical Investigations of Hand-held Scanning Systems (Update) View project Determination of Intensity-Based Stochastic Models for Terrestrial Laser Scanners Utilising 3D-Point Clouds View project Untersuchungen zum Genauigkeitspotential des ter-restrischen Laserscanners Leica BLK360, 2018. https://www.researchgate.net/publication/325346976 (accessed September 18, 2020). [57] S. Soudarissanane, R.C. Lindenbergh, R. Lindenbergh, B. Gorte, Reducing the error in terrestrial laser scanning by optimizing the measurement set-up, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. XXXVII (2008) 615–620. http://www.deos.tudelft.nl/ (accessed January 21, 2021). [58] M. Attenni, Informative Models for Architectural Heritage, Heritage. 2 (2019) 2067–2089. https://doi.org/10.3390/heritage2030125. [59] C. Santagati, M. Lo Turco, From structure from motion to historical building information modeling: populating a semantic-aware library of architectural elements, J. Electron. Imaging. 26 (2016) 011007. https://doi.org/10.1117/1.jei.26.1.011007. [60] T. Messaoudi, P. Véron, G. Halin, L. De Luca, An ontological model for the reality-based 3D annotation of heritage building conservation state, J. Cult. Herit. (2017). https://doi.org/10.1016/j.culher.2017.05.017. [61] Innovmetric, The Smart 3D Metrology Digital Ecosystem Resumen de productos, (2021). https://www.innovmetric.com/es/products/products-overview (accessed April 5, 2021). [62] 3D Systems, Geomagic Studio, (2015). http://www.3dcadportal.com/geomagicstudio.html (accessed March 4, 2020). [63] Artec 3D, Geomagic Wrap | Download Free Trial | Artec3D, (2021). https://www.artec3d.com/3d-software/geomagic-wrap (accessed April 5, 2021). [64] Geo-Plus, VisionLidar - Point Cloud processing software - Scan to BIM, (2020). https://www.geo-plus.com/point-cloud-software/ (accessed March 29, 2020). [65] S. Garagnani, GreenSpider - The Autodesk Revit point clouds plugin - TC Project, (2015). http://www.tcproject.net/pivotx/?p=greenspider (accessed January 30, 2021). [66] B. Systems, Pointools | Point Cloud Processing Software | Bentley, (2020). https://www.bentley.com/en/products/product-line/reality-modeling-software/bentleypointools (accessed January 30, 2021). [67] S. Garagnani, A.M. Manferdini, Parametric accuracy: Building information modeling process applied to the cultural heritage preservation, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. - ISPRS Arch. 40 (2013) 87–92. https://pdfs.semanticscholar.org/1682/3b3e272038fec08b38ea9267588efbad5589.pdf% 0Ahttp://www.scopus.com/inward/record.url?eid=2-s2.0- 84922519650&partnerID=tZOtx3y1. [68] J.E. Nieto-Julián, L. Lara, J. Moyano, Implementation of a TeamWork-HBIM for the Management and Sustainability of Architectural Heritage, Sustainability. 13 (2021) 2161. https://doi.org/10.3390/su13042161. [69] D. Rebolj, Z. Pučko, N.Č. Babič, M. Bizjak, D. Mongus, Point cloud quality requirements for Scan-vs-BIM based automated construction progress monitoring, Autom. Constr. 84 (2017) 323–334. https://doi.org/10.1016/j.autcon.2017.09.021. [70] V. CAFFI, A. PAVAN, M.A. PIGNATARO, Il processo edilizio supportato dal BIMM: l’approccio InnovANCE / BIMM Enabled Construction Pprocesses: the InnovANCE Approach, INNOVance BIM Object Ital. Build. Ind. (2014) 91–109. https://re.public.polimi.it/handle/11311/842140#.YA6OYuj0mUk (accessed January 25, 2021). [71] B. Forum, Level of Development Specification , BIM Forum, (2013). 37 https://bimforum.org/lod/ (accessed January 25, 2021). [72] P. Maiezza, AS-Built reliability in architectural HBIM modeling, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. XLII-2/W9 (2019) 461–466. https://doi.org/10.5194/isprsarchives-XLII-2-W9-461-2019. [73] H. Macher, T. Landes, P. Grussenmeyer, From point clouds to building information models: 3D semi-automatic reconstruction of indoors of existing buildings, Appl. Sci. 7 (2017) 1–30. https://doi.org/10.3390/app7101030. [74] R. Quattrini, E. Malinverni, From TLS to HBIM. High quality semantically-aware 3D modeling of complex architecture, … Sens. Spat. …. (2015). http://search.proquest.com/openview/096844567088e33d0d516f1ced6ccc43/1?pqorigsite=gscholar&cbl=2037674 (accessed April 21, 2016). [75] R. Quattrini, R. Pierdicca, C. Morbidoni, Knowledge-based data enrichment for HBIM: Exploring high-quality models using the semantic-web, J. Cult. Herit. (2017). https://doi.org/10.1016/j.culher.2017.05.004. [76] L. Barazzetti, F. Banfi, R. Brumana, G. Gusmeroli, M. Previtali, G. Schiantarelli, Cloud- to-BIM-to-FEM: Structural simulation with accurate historic BIM from laser scans, Simul. Model. Pract. Theory. 57 (2015) 71–87. https://doi.org/10.1016/j.simpat.2015.06.004. [77] A. Georgantas, M. Brédif, M. Pierrot-Desseilligny, An accuracy assessment of automated photogrammetric techniques for 3D modelling of complex interiors, Int. Arch. Photogramm, Remote Sens. Spat. Inf. Sci. XXXIX-B3 (2012) 23–28. [78] G. Teza, A. Pesci, A. Ninfo, Morphological Analysis for Architectural Applications: Comparison between Laser Scanning and Structure-from-Motion Photogrammetry, J. Surv. Eng. (2016). http://ascelibrary.org/doi/abs/10.1061/(ASCE)SU.1943-5428.0000172 (accessed April 28, 2016). [79] L. Jurjević, M. Gašparović, 3D data acquisition based on OPENCV for close-range photogrammetry applications, Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. XLII- 1/W1 (2017) 377–382. https://doi.org/10.5194/isprs-archives-XLII-1-W1-377-2017. [80] A. Koutsoudis, B. Vidmar, G. Ioannakis, F. Arnaoutoglou, G. Pavlidis, C. Chamzas, Multi-image 3D reconstruction data evaluation, J. Cult. Herit. 15 (2014) 73–79. https://doi.org/10.1016/j.culher.2012.12.003. [81] C. León-Robles, J. Reinoso-Gordo, J. González-Quiñones, Heritage Building Information Modeling (H-BIM) Applied to A Stone Bridge, ISPRS Int. J. Geo-Information. 8 (2019) 121. https://doi.org/10.3390/ijgi8030121. [82] J.E. Nieto-Julián, D. Antón, J.J. Moyano, Implementation and Management of Structural Deformations into Historic Building Information Models, Int. J. Archit. Herit. (2019) 1–14. https://doi.org/10.1080/15583058.2019.1610523. [83] W. Mukupa, G.W. Roberts, C.M. Hancock, K. Al-Manasir, A review of the use of terrestrial laser scanning application for change detection and deformation monitoring of structures, Surv. Rev. 49 (2016) 1–18. https://doi.org/10.1080/00396265.2015.1133039. [84] D. Lague, N. Brodu, J. Leroux, Accurate 3D comparison of complex topography with terrestrial laser scanner: Application to the Rangitikei canyon (N-Z), ISPRS J. Photogramm. Remote Sens. 82 (2013) 10–26. https://doi.org/10.1016/j.isprsjprs.2013.04.009. [85] M. Kazhdan, H. Hoppe, Screened poisson surface reconstruction, ACM Trans. Graph. 32 (2013) 1–13. https://doi.org/10.1145/2487228.2487237. [86] P. Arias, C. Ordóñez, H. Lorenzo, J. Herraez, Methods for documenting historical agroindustrial buildings: a comparative study and a simple photogrammetric method, J. Cult. Herit. 7 (2006) 350–354. https://doi.org/10.1016/J.CULHER.2006.09.002. [87] X. Yang, M. Pierre Grussenmeyer, M. Andreas Georgopoulos Professeur, N.-G.M. 38 Pierre Drap, M. Monica Siroux, ÉCOLE DOCTORALE MSII (ED n°269) LABORATOIRE ICUBE (UMR 7357) THÈSE prsente par : 3D modeling of built heritage: from geometric models to HBIM THÈSE dirige par : RAPPORTEURS : AUTRES MEMBRES DU JURY, Strasbourg, 2018. http://www.theses.fr/2018STRAD054 (accessed January 29, 2021). [88] A. Ghaffarianhoseini, J. Tookey, A. Ghaffarianhoseini, N. Naismith, S. Azhar, O. Efimova, K. Raahemifar, Building Information Modelling (BIM) uptake: Clear benefits, understanding its implementation, risks and challenges, Renew. Sustain. Energy Rev. 75 (2017) 1046–1053. https://doi.org/10.1016/j.rser.2016.11.083. [89] Z.Q. Liu, F. Zhang, J. Zhang, The building information modeling and its use for data transformation in the structural design stage, J. Appl. Sci. Eng. 19 (2016) 273–284. https://doi.org/10.6180/jase.2016.19.3.05.