Digitalization based on high-resolution scanning and HBIM tools for damage assessment of the José de Alencar house

Heritage buildings are of great importance to the human perception of the culture of a community. HBIM (Historic Building Information Modelling) tools offer a possibility of an improved data set of information related mainly to the restoration and preservation of historical buildings. This work aims to assess the damage to the historic house by employing integrated HBIM tolls and experimental procedures. The historic house was evaluated by visual inspection of the historic house, 3D modeling with REVIT, and 3D modeling based on point cloud data. The comparison between the two 3D modeling techniques showed a level of damage consisting of a difference between the levels of the roofs. In addition, the visual inspection detected cracks in the walls which agrees with the damage observed from the 3D model’s comparison. Results indicate that HBIM tools significantly contribute to damage assessment in heritage constructions.


Introduction
The existence of heritage constructions (HC) in contemporary society is of great relevance due to its contribution to the cultural aspects of a community. HC is the physical representation of part of the history of human development. Historic buildings provide evidence of past societies and reflect their lifestyle, social organization, and advances [1]. Preserving historic buildings is a complex task that sometimes needs to balance the costs and prioritize the interventions. When maintenance is regularly adopted in construction early, it is possible to decrease between 40 to 70% of the total intervention costs [2]. In addition, also more dramatic interventions can be avoided when continuous maintenance and monitoring are performed [3]. As part of the valorization of these historical buildings, the development of techniques for supporting the maintenance, as the documentation, is necessary [4,5].
While heritage maintenance should involve a complete comprehension of the buildings concerning constructive processes (materials, historical importance, compatible technologies, and all possible documentation of the building characteristics) the documentation can be a complex activity because it can involve beyond written documents, the interpretation of geometrical components of the cultural object assessed [6,7]. In this sense, historic documentation plays an essential role in maintenance measures, also being important to include newly available technology. Digital scanning, for instance, can contribute to the documentation of HC by allowing the collection of minor details in a 3D perspective, changing the perspective of the registers, and even contributing to salving the originality of a monument. Another advantage of digital technologies is that a considerable range of tools, including smartphones, digital cameras, manual scanners, and drones, can be integrated into the process of data capture. Nonetheless, the observation of the HC characteristics is necessary to decide on the strategy of data collection. A column with a considerable quantity of artistic details, for instance, requires a more detailed and considerable density of the points cloud than a single wall. However, digital tools also present limitations and issues that must be 1 3 30 Page 2 of 16 addressed in further developments, such as the data amount. A digital model can easily achieve 50 Gb of extension, and that is a limitation to the digitalization of cultural heritage.
Heritage documentation systematically collects and archives tangible and intangible elements of historic structures and environments [8]. The documentation provides accurate information for the correct building's conservation, monitoring, and maintenance [9]. Techniques for acquiring modern construction documentation may include 2D/3D CAD (Computer Aided Design). Also, 3D modeling improved with parametric and feature-based modeling tools with a level of intelligence in model elements. The recent development of the concept of BIM (Building Information Modelling) incorporates the main developments in 3D modeling, including parametric and feature-based modeling combined with a dynamic 3D database for storing information relating to buildings [8].
Beyond 3D modeling, BIM tools assemble a cooperative environment where the material and structural properties, maintenance history, interventions, and other relevant information of the historic construction can be put available in tridimensional representation (4D). If the 3D model is implemented continuously, it can also be valuable for management over time, with evolution to costs and for the life cycle assessment, and further to the direction of a digital twin [10,11]. Applying these new technological tools in the rehabilitation, restoration, documentation, and maintenance of cultural buildings is very advantageous because management, planning, and sustainability data can be integrated, and all data can be available in the same place [12][13][14].
In recent years, several studies have been developed using the BIM platform, attracting the scientific community's attention for application in restoration design and management of the facilities [15]. In this context, [16] developed a management tool based on BIM with a multidimensional model for solving inconsistencies in integrating the information of a structural health monitoring system. The wood structures of the Feiyun Pavillon in Shanxi, China, were used to demonstrate the integration working. In the same scenario, [17] consider that the modeling, the formulation of the database, and the management of the structures information can offer several benefits if implemented in the activities of operation and maintenance of a construction. However, the authors highlight that specific literature is necessary for practical cases in historic buildings. In this line, reports on the field's techniques and strategies for data acquisition are also necessary [18]. According to [19], detailed data collected with advanced geomatic technologies may accurately reproduce all singularities and specifics of the objects commonly used to collect the geometric characteristics of architectural heritage. In the case of historic buildings, the virtual reconstruction procedure of historical-cultural heritage is not easy, as the object-to-model consists of components whose heterogeneous, complex, and irregular characteristics and morphologies are not represented in the BIM software libraries [12].
Introduced by [20], HBIM (Historic Building Information Modelling) is a complex model composed of a library of parametric objects with tools capable of mapping smart objects to survey data [21]. Proposed data integration of laser scanners and images named HBIM, involving reverse engineering, where interactive parametric objects representing the architectural elements are mapped using of laser scan or photogrammetric survey. The authors observe that HBIM pursues the modeling documentation of architectural elements according to artistic, historical, and constructive typologies. In addition, the HBIM library begins with the 3D modeling of parametric objects, employing point cloud integration, photogrammetry, historical documentation, and Graphisoft Archicad software, according to [22] and [23]. The development of HBIM mainly focuses on the BIM methodology for the historic building features and on increasing the geometric accuracy of the model data enrichment [24]. [25] reports a strategy for HBIM that aims to support the planning and activities of maintenance and conservation of historic buildings. The flowchart is divided into five steps: model planning, data collection, geometric survey, structure detailing, and HBIM modeling. According to the authors, the proposal is flexible and recursive because it adapts to information available along with the work development.
Digital scanning is the first step to creating smart models for historic buildings [26]. Surveying digital techniques available to acquire data needed to generate accurate as-built Building Information Models may include terrestrial laser scanning (TLS) and photogrammetry. [27] and [28] consider as main steps for digital scan: the survey of spatial and documentation data, the processing and generation of the point clouds, and the BIM modeling created from the point clouds. Photogrammetric techniques use images taken from different viewpoints to record the 3D geometry of a building, becoming very popular, especially for recording e, especially for cultural heritage applications [8] TLS and Photogrammetry techniques require several pre-processing steps to create 3D CAD and BIM models.
For instance, [29] studied various data processing workflows for three different applications based on low-cost data acquisition techniques. Initially, as-built BIM for a historic building based on a registered point cloud according to Ground Control Points was performed. Then, structural analysis was applied to a damaged bridge using Finite Element analysis software. Finally, parametric automated feature extraction from captured point cloud for reverse modeling and fabrication was used. The authors observed that all techniques present high efficiency in documenting architectural heritage.
Nowadays, approaches reported in the literature are developing studies to automate the processes of segmentation, classification, and modeling of point clouds resulting in more accurate three-dimensional surfaces of building elements [30,31] and to support an approach of damage interpretation and structural assessment based on a machine applying Semantic Web technologies [32], respectively. Works involving capturing large-scale data from facades using TLS [33] are also being monitored, as the comparison of precision between low-cost and easy-to-use scanners, and state-of-the-art laser equipment, in the acquisition of point clouds of buildings. [34] and the exploration of a 3D parametric model as a basis for an HBIM methodology to create a finite element model (FEM). Interoperability issues and difficulties in the Scan-to-HBIM process were investigated [35].
Finally, in the same line of research as this work, [3] conducted a comparative analysis between the parametric model and the real geometry of a gantry of an HC dating from the 18th century. Data collection was carried out using the TLS technique, and the structural element was modeled using the Revit software. The structural deviations of the frame were measured using Dynamo©, and CloudCompare tools were used to validate the results. The results showed that the most unfavorable constructive unit deviations were between 0.5 mm and 11.7 mm, consolidating the technique as adequate to review visual records and analyze structural deviations.
Thus, this work performed the damage mapping of a historic house integrating 3D modeling HBIM tools with point cloud data and Revit Autodesk software. An experimental procedure of visual inspection and numerical analysis using the finite element method supported the results of this work.

Case study: José de Alencar house
Located in the neighborhood of Messejana, in the city of Fortaleza, State of Ceará in Brazil, the house of José de Alencar, built in the 19th century, is currently a cultural center belonging to the Federal University of Ceará, located in the Site of Alagadiço Novo. In 1964 the house was considered by the National Historical Heritage Institute-IPHAN a historical heritage. In addition to the house, the site has the ruins of an old mill that worked there.
The house has important symbology because it was the home of the greatest Brazilian novelist, the writer José de Alencar, born in 1829 and a symbol of Brazilian literature and politics. The books entitled "Iracema" and "O Guarani" are some famous works from José de Alencar. In the Brazilian scenario, the writer stood out mainly due to his novels with more nationalist and patriotic themes, getting out of the European model of writing, which until then was widely used, especially for its regional writings. The expressive writer received as tribute a theatre and a square in the city of Fortaleza with his name, and in Rio de Janeiro city, a statue was done in his honor in the neighborhood of Flamengo.
The walls of the historical building consist basically of solid ceramic brick joined by a layer of mortar based on lime, sand, and clay. The covering structure comprises Carnaúba (Copernicia prunifera) woodwork and ceramic tiles. The house, with a total area of 88.29 m 2 , is divided into three rooms and a balcony, representing the typology of construction very well at the time.
The house presents a simple country architecture, preserved in its original environment, a domestic and rural atmosphere that, as a historical heritage, can be recognized for its symbolic, artistic, landscape, and cultural value, richly showing the cultural and natural aspects of Ceará. Figure 1 depicts two façades José de Alencar house.

Material and methods
The HBIM implementation consists of several steps according to the existing documentation and information of the historic building, the technology used, and the level of obtained data required. This HBIM workflow involves three main stages: planning, acquisition and processing, and BIM modeling. Figure 2 presents the H-BIM procedural workflow of this work.

Planning
The planning is the first action to start the HBIM project implementation. This stage involved two other steps in this work: the objective definition and the choice of equipment. A step that can be included in the planning and precede the objectives and choices of equipment is historical and archival research, which can provide historical data on the evolution of the building, such as renovations, change of owners and purposes of use, and architectural influences, among others. This phase can also return information about regulatory requirements, restrictions, and specific needs that the building may present [36].
The objective definition consisted of defining the H-BIM project target, precisely the comprehension of the project needs. So, the main goal was to create a digital 3D model of the José de Alencar house using the available technology based on 3D laser scanning. The architectural typology of the house and the 3D model desired defined the choice of equipment. A laser scanner Leica BLK360 was used for the scanning process due to the high precision in the survey as the historic building was studied. The planning also may include the level of accuracy and reach, the ways of storing and acquiring data, and the definition of scenes and targets.

Data acquisition and processing
The phase of acquisition and processing involves three main steps: data acquisition, pre-processing, and 3D model processing. The data acquisition is part of the fieldwork, which may include sketches, direct measurements, an approximation of the object, and a survey of the control point. Initially, the data acquisition was planned to verify details of the building, such as the presence of doors and windows and unique structural and architectural elements with the need for special detailing. Thus, due to the architectural simplicity of the construction and no special details identified, the process of scanning was planned in other steps.
To correctly scan with a Leica BLK360, a house with many rooms is necessary to scan inside each room, door, and corridor, among other building spaces. Besides, to ensure that the laser scanner has the line of vision of all elements to be captured. This means planning a better place to configure the tripod of the equipment. From the 18 scenes (stations), 4 were recorded inside the house, and 14 were recorded outside the building. The images must be taken with a good overlap between them according to the sequence to link all pictures in one only point cloud during the processing. Figure 3a presents the tripod configuration to the laser scanner, and Fig. 3b shows the stations' location (points in red) and the linked scenes (green lines).
The pre-processing phase consisted of three main stages: registering the digitalization, fitting the linked images in the planimetric and altimetric levels, and cleaning the point of the clouds according to the needs of the  project. Cyclone REGISTER 360 ® was used to register the digitalization (Fig. 4a). Each station and link is adjusted in the planimetric (Figs. 4b and d) and altimetric (Figs. 4c and e). Each image, according to the linking and sequence of the scenes, is adjusted to assure the accuracy of the modeling. The photos are overlapped at each point until they reach the correct position on each image as assembling a puzzle game, where so many pieces need to be joined to generate only one object.
The pre-processing usually consumes half of the acquisition time on site due to so many steps of data treatment. The point cloud is also cleaned, selecting points to be excluded. In the data of the house, external points of images of trees (Fig. 5a) were selected and discarded from the point cloud. Due to the characteristic of historical heritage, it was defined not to clean the part of the trees around the building (Fig. 5b). The pre-processing also may involve the reduction of noises and the optimization of the clouds. Revision processing involved using the TruSlicer function, where each color (Fig. 5c) represents the points of one scene.
The 3D model processing is the last step of the final generation of the 3D point cloud, storing and exporting files. In this work, the files were exported as LGS, which can be visualized in applications such as TruView or JetStream Viewer, and files rcp/rcs that allow the data transfer for software such as AutoCAD and BIM. Figure 6 depicts the 3D model ready for exportation with revised liking and cleaning of the cloud.
The quality control is also verified in this stage, including accuracy verification. The overall quality of the scanning measured is a bundle error of 0.010 m, an overlap of 31%, a strength of 53%, and cloud-to-cloud 0.010 m, considering 18 scenes and 22 link count. Figure 7 presents the high quality of the point cloud generated at the José de Alencar house.

Point cloud assembly and BIM modeling
The 3D model of the José de Alencar house was developed using the software REVIT Autodesk (Fig. 8a), based on the existing architecture available in AutoCAD 2D provided by National Historic and Artistic Institute (IPHAN), as presented by Fig. 8b. The modeling developed from the floor plan, section, façade, and roof plan (AutoCAD 2D) used tools such as walls, floor, roof, sloping glazing, doors, and windows. The point cloud was assembled in the Revit software. The point cloud file was converted in RCP extension to integrate the point cloud with the REVIT using the Recap Pro Autodesk. The overlapping of 3D models (point cloud and Revit) generated the geometric measurement of the house. Figure 9 presents the 3D model of the façades frontal and lateral of the house.
In addition to the steps belonging to the HBIM workflow, visual inspection techniques and structural numerical analysis of the house were included in this work to contribute to the analysis and results.

Visual inspection
Visual inspection, as a non-destructive technique, is usually the primary method to characterize the actual condition of a historic building. A visual inspection of the José de Alencar house was performed to assess the current state of conservation. The structural system, the walls, the floor, and the covering were visually observed. The inspection covered the presence of cracks, the damage to the structure, the type of materials, and the constructive process of the walls and floor. Besides the environmental condition of the construction site and elements close to the external part of the house was inspected. The visual inspection information is very important to validate the level and typology of damage obtained from the digital data analysis comparison. Pictures of the damages (cracks, deterioration of the floors, structure, and covering) were taken to produce a photography record of the inspection. In addition, relevant details of the characteristics of the house were documented, such as cracks, masonries tilting and moisture presence.

Numerical simulation
In this analysis, the macro-modeling technique was used for the main constituent element of the structure, the masonry of solid ceramic bricks. This technique represents elements, connections, and an interface between them in a continuous equivalent way. The so-called macroelement, which has generalized forces and deformations to describe the behavior of these different materials in a simplified model, represents the structure as a whole.
The 3D model of the structure was designed in REVIT ® Autodesk software and analyzed in Ansys ® software version 17.1. The MultiZone function was used to discretize the mesh of the 3D mode generating a hexagonal mesh. The element used was HEX20, which has 6 nodes and 3 degrees of freedom. Altogether, 10,996 elements and 67,526 nodes were obtained. Figure 10 shows the mesh used for modeling.

Loadings and mechanical properties of the model
For the loads acting on the structure, values for the loads from the roof were searched in the literature. The load used to represent the roofing, composed of a carnauba structure and colonial tiles following [37], was 0.85 KN/m 2 on the side masonry. The overload used was 50% (0.425 KN/m2) totaling a load of 1.275 KN/m 2 [38].
The mechanical properties of the macroelement were taken from the literature, with the Modulus of elasticity (E)   [39], the Specific weight (w) and the Compressive strength (Fc) obtained by parameters estimated in [40] and the Poisson coefficient (v) [38,39,41], as can be seen in Table 1.
The hypothesis of damage to Jose de Alencar House assumes that the walls suffered vertical displacements. Therefore, in the numerical simulation, these displacements were made in order to try to simulate reality, as represented in Fig. 11. The lower faces of the walls were set in place, except for those that were displaced, to simulate the effect of the residence's foundations.

Point cloud assembly and HBIM 3D modeling
The 3D point cloud and the 3D model of the José de Alencar House in REVIT are shown in Fig. 12.
The results (Fig. 12) show the benefits of both technology, laser scanner, and BIM a very suitable solutions for modeling and managing information related to existing heritage buildings. From the 3D models, the architectural  geometry is depicted, allowing a large range of documentation and analysis of the house. The obtained models also present a high level of accuracy regarding the constructive details of the house, like the masonry geometry information.

Comparison between 3D Models
The comparison between the 3D models (point cloud and Revit) allowed the assessment of divergence in the geometrical characteristics of the house. The two roofs presented a level of difference: of the balcony (Fig. 13a) and the main roof of the house (Fig. 13b). Notice that the model generated from the laser scanner was obtained in 2021, whereas the model built in Revit was based on an existing architecture register from 2002 provided by IPHAN. The findings of the unevenness between the two coverings are attributed to the accuracy of the modeling for damage detection. Overlapping the models is observed that the structure of the house may be under deformation over time.
Considering that the house is composed of walls of solid bricks joined with a layer of masonry and presented no concrete structural elements, such deformation may indicate a level of movement of the walls which compose the structure of the building. It can be highlighted that in a structural interpretation, the house walls are under compression due to their weight and the weight of the covering. Another possibility may be load-settlement behavior due to the interaction between soil and structure. So, the 3D models offer many options for assessing the occurrence and typology of damages in historic buildings representing a valuable tool for detecting damage and planning to retrofit.
Wall thicknesses found in the point cloud and floor plan (Fig. 14) models of the existing architectural design were also analyzed. The comparison between the models returned   , including in the course of its structure, which is the case of wall P1 (437.00 mm, 428.00 mm and 423.00 mm). These variations are justified because the building was built during the nineteenth century without a precise construction process and skilled labor, in addition to possible renovations that were carried out over time. The previous 2D records of the Jose de Alencar house took the wall thickness as an average measure, and not as an element with minor variation.

Damage mapping
Results obtained by visual inspection of the damage of the José de Alencar house are shown in Fig. 15. The presence of cracks detected in the walls is depicted in Fig. 15a, whereas Fig. 15b presents the movement of an inner wall and the lowering of the floor level of the room. Characterizing damage mapping in masonry structures allows the assessment of causes, effects, and consequences.
The presence of vertical and inclined cracks was observed (Fig. 15a). Such deterioration may be attributed to land movements, as the cracking patterns are following the movement of the inner wall (Fig. 15b). Cracks detected in the walls presented in general, a diagonal direction of propagation, with angles of 45° and a plane surface. Besides the movement of the inner wall, the occurrence of cracks may be a consequence of a possible elevation of the roots of a specimen of the tree (Mangifera indica) located close to the building.
The generalized presence of cracks in the walls may indicate that the distribution of tensions in the masonry walls is associated with the lack of rigid structural elements for better distribution of loads. Such a level of tension may lead to the continuous evolution of crack propagation and crack widths, compromising the integrity of the historic house. These results agree with the difference in the levels of the roof observed in the comparison of both 3D models of the house (see Sect. 4.2).

Numerical analysis
After the numerical analysis of the structure, it was observed that a large part of the building is under compressive stresses that vary between 0 and 1.66 MPa and come from the selfweight and the roof. The maximum compressive stress values found are below the resistance values used by the macro model. Therefore, there is no danger of compression failure.
Due to their susceptibility to tensile stress, there are also tensile zones near the frames. Tensions were found where there were cracks in the real structure, reaching values of 6.86 MPa and reinforcing the hypothesis that there were earth movements that caused the structure to suffer displacements along with the soil (as seen in Fig. 16 and Fig. 17). The tensile stress values found exceed the resistance value attributed to the macroelement. Therefore It was possible to analyze the real cracks and those manifested in the model, which guarantees the calibration of the simulation. Thus, Fig. 18 expresses the comparison according to the numerical modeling results. It can be observed that in Fig. 18(a) and (b), the tensions that caused the panels to move, according to the presented boundary conditions, loads, and mechanical properties of the model, converged to the real positions on the wall that they are. A similar fact happened in the analysis expressed by Figs. 18(c) and (d). Figures 18(e) and (f) show the stress concentration region that consequently started the crack at this point.

Conclusions
Through the case study performed of the José de Alencar house, it is possible to notice that H-BIM tools allow the assessment of damages in historic buildings. So, the following conclusions may be derived from the work: • The laser scanner survey obtained dimension information and data to build a 3D model of the historic house. The point cloud generated presented high accuracy of all elements of the house, allowing the assessment of the real condition of the building.

Data availability
The authors confirm that the data supporting the findings of this study are available by email request to emesquita@ufc.br.

Conflicts of interest
The authors declare that there are no competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.