Digital transformation through multi-device HBIM workflow: a case study on supporting the adaptive reuse of the Odd Fellows Building in Atlanta, Georgia

This paper presents a multi-device scan-to-HBIM workflow developed for the adaptive reuse of the Odd Fellows Building (b.1911), a socially significant historic landmark with seven stories above ground and a basement, located in Atlanta’s Sweet Auburn National Historic Landmark District. As a major contributing structure within the culturally significant corridor, the building’s adaptive reuse demands precise documentation, rapid data collection, and efficient stakeholder communication. To meet these goals within strict time and funding limits, the research integrates Terrestrial Laser Scanning (TLS), Matterport 360-camera technology, and Unmanned Aerial Vehicles (UAVs) to create an efficient and accessible pipeline for documenting historic buildings, assisting with renovation and construction processes. TLS is employed as a benchmark capture method to establish a baseline point cloud for aligning all other data. Matterport technology enables fast interior scanning while providing high-resolution visuals and an interactive virtual tour platform. UAVs are deployed to supplement data collection in hard-to-access exterior areas, especially where ground-based reality capture tools encounter spatial or logistical barriers. This not only accelerates data acquisition under time and budget constraints but also offers stakeholders the ability to remotely revisit the space without geographic limitation. This multi-device strategy optimizes the trade-off between capture speed and spatial tolerance, accommodating a complex mix of preserved historic elements (exterior and lobby space) and altered upper floors. Besides, a full TLS capture approach could cause potential data saturation and processing inefficiencies. The multi-device approach has multiple data processed registered to provide one combined point cloud, which is integrated into a semantically rich HBIM model developed in Autodesk Revit at Level of Development (LoD) 300. Although not a full Digital Twin (DT), the resulting model establishes a foundational framework for future digital twinning and immersive applications. By demonstrating how different reality capture technologies can be effectively integrated to accommodate site limitations, stakeholder needs, and funding constraints, this case study contributes to the broader discourse on digital transformation in construction, particularly the evolving intersection of BIM, remote visualization, and the early stages of DT development in the preservation sector.

1 Introduction

The architecture, engineering, construction, and operations (AECO) industry is undergoing a profound digital transformation, driven by the convergence of Building Information Modeling (BIM), immersive visualization platforms, and reality capture technologies. Within this transition, Heritage Building Information Modeling (HBIM) has emerged as a crucial workflow for documenting, managing, and reusing heritage structures. Unlike conventional BIM practices that are oriented toward new construction, HBIM processes must contend with the complexities of irregular geometries, incomplete historical records, and the need to balance conservation ethics with renovation imperatives. Recent advances in terrestrial laser scanning (TLS), unmanned aerial vehicles (UAVs), and photogrammetry have expanded the potential of HBIM by enabling high-fidelity data capture. Traditional on-site surveys rely on human labor measurement, which is time-consuming and labor-intensive. Current HBIM methodology applies digital reality capture tools such as TLS, UAVs, etc., to achieve more accurate and time-saving measurements (Yang et al., 2020). However, despite these technological developments, challenges remain integrating diverse capture devices into efficient, accessible, and resource-conscious workflows.

Recent studies have shown that HBIM research has expanded rapidly over the last decade, shifting from a primary focus on geometric representation to a broader analysis of information management, conservation workflows, and integration with other digital systems. Yang et al. (2020) reviewed built heritage modeling and emphasized the need to integrate HBIM with photogrammetry, GIS, and semantic enrichment techniques to support more robust documentation and analysis. Dore and Murphy (2017) emphasized that the complexity of heritage structures requires adaptive workflows that balance data density and semantic modeling accuracy. Inzerillo et al. (2025) explored AI-powered modeling for heritage bridges via semantic segmentation in point clouds to reduce modeling time and improve geometric quality. Zhang and Zou (2022) used a CiteSpace-based analysis to map research hotspots and trends in HBIM, finding a change toward preventive conservation, multi-source data integration, and lifecycle management of built heritage assets. Parente et al. (2025) reviewed most recent HBIM literature, which focuses on information management and the integration of non-geometric data, such as diagnostic, monitoring, and conservation records, into BIM environments, underscoring ongoing challenges in interoperability and data structure. Alshawabkeh and Baik (2023) demonstrated how combining laser scanning and photogrammetry can enhance scan-to-HBIM workflows by enriching point clouds with texture and damage information, while Giuliani et al. (2024) proposed an HBIM pipeline for large-scale architectural heritage that emphasizes fit-for-purpose workflows and interoperability. In recent years, the convergence of HBIM and Digital Twin (DT) frameworks has gained increasing attention as scholars and practitioners seek to transform static 3D heritage models into dynamic systems capable of monitoring and simulation. There is the potential of linking HBIM with real-time data and monitoring systems but also the complexity of doing so in practice is very challenging (Lucchi, 2023). Niccolucci and Felicetti (2024) introduced an ontology-based approach that connects HBIM models with sensor networks and cultural heritage data structures, demonstrating how real-time environmental inputs can enrich the semantic layers of a digital twin. Extending these efforts, Yu et al. (2025) provided a systematic review of digital technologies for architectural heritage risk management, identifying HBIM–DT integration as a crucial pathway toward predictive maintenance and preventive conservation. These studies underscore the research interest shift from geometric documentation toward intelligent, interoperable heritage information systems.

The reuse of historic structures places particular demands on digital documentation, as it normally needs an initial building survey to help stakeholders understand the overall structure and the building’s as-builtconditions (Khalil and Stravoravdis, 2022). In addition, having digital models, especially BIM models, can accelerate design proposals and streamline construction progress. Adaptive reuse projects can face compressed timelines, constrained budgets, and site limitations that make conventional building survey approaches impractical. Moreover, heritage sites present additional layers of significance: beyond their material fabric, they embody cultural, social, and historical narratives that must be studied and safeguarded in the renovation process. In this context, there is a growing demand for utilizing different reality capture tools survey and record for historic structures. In addition, historic building practitioners often face various challenges in project management, financial loss, and project delay during the restoration process due to the lack of digital documentation and updated digital models. Traditionally, information about historic structures has been fragmented across different formats and managed independently by different specialists. HBIM offers a unified digital environment that can overcome these long-standing challenges by improving interoperability, information integration, and interdisciplinary collaboration (Khan et al., 2022). The alignment of HBIM with the early stages of DT development further underscores the importance of scalable digital pipelines that support not only documentation but also long-term stewardship, stakeholder engagement, and immersive applications (Cassar et al., 2025). Despite this growing body of research, relatively few studies documented the combination of multiple capture devices under strict time and budget constraints in adaptive reuse projects. The presented case study addresses this gap by testing a rapid, multi-device HBIM workflow that is calibrated to the availability of limited resources while remaining sensitive to both critical heritage values and stakeholder needs.

This paper responds to these challenges by proposing and testing a multi-device HBIM workflow for adaptive reuse. Specifically, it investigates how TLS, Matterport scanning, and UAV photogrammetry can be strategically combined to overcome site access limitations, mitigate data saturation, and deliver integrated outputs within severe time and funding constraints. The approach is evaluated through the case study of the Odd Fellows Building, a culturally significant landmark in Atlanta’s Sweet Auburn Historic District. Once a hub of African American professional and social life, the structure now faces both physical deterioration and the opportunity of renewed community service through its adaptive reuse project. By applying a multi-device digital strategy, this study demonstrates how reality capture technologies can be orchestrated to produce a semantically rich HBIM model at Level of Development (LoD) 300, serving immediate renovation needs while laying a foundation for future digital twinning.

While previous studies combined different reality capture devices for built heritage documentation, the innovation of this work lies in developing a significance-driven and resource-constrained multi-device HBIM workflow. The proposed framework prioritizes reality capture devices based on the cultural and historic importance of building components, balancing accuracy and efficiency according to project significance and accessibility. This study is also distinguished by its real-world application of an adaptive-reuse project under a 1-day scanning and 1-week modeling limit, demonstrating a pragmatic model for limited-budget preservation project. The contribution therefore extends beyond technical integration to provide a decision-making framework that can be adaptable to other built heritage facing similar financial and logistical limitations. Through this contribution, the paper situates HBIM not merely as a technical tool but as an integral part of the broader digital transformation in construction. It highlights the potential of hybrid capture workflows to support both project-specific requirements and long-term cultural heritage preservation, bridging the gap between rapid documentation and comprehensive digital integration.

2 Materials and methods

Creating a HBIM model through reality capture has been demanded urgently to assist stakeholders’ renovation and demolition proposal, such as producing two-dimensional drawings and renderings. Meanwhile, under short notice and limited working-time windows, applying reality capture tools through a scan-to-BIM process significantly reduces on-site time and minimizes human error compared to manual surveys (Rocha et al., 2020). From long-term preservation perspective, establishing a digital model would also be beneficial to keep archive integrity and assist future renovation and maintenance needs. In order to achieve HBIM model creation goal, selecting appropriate reality capture tool to use becomes challenging. Different reality capture tools have different advantages and disadvantages. The selection of reality capture tools should depend on project goals, site limitation, and resource availability to make a balanced decision.

As the ultimate product goal of creating HBIM model, the reality capture devices that can produce point cloud or assist HBIM process are included in this discussion: TLS, UAVs, close-range camera (CRC), Matterport, mobile handheld scanner (MHS), and total station (TS). To align with the balanced decision goal, those devices are categorized by project goals and site limitations to demonstrate their feasibility under different scenarios.

The project goal represents diverse use cases and different LoD for the final product that the involved stakeholders try to achieve. Table 1 summarizes how different reality-capture devices align with specific project goals, including point-cloud generation, high-resolution visuals, and geo-referencing accuracy. The specialty of Matterport offering a virtual tour is valuable, which can provide remote access to stakeholders without physical geolocation restrictions. Meanwhile, it provides high-resolution visual data so that any people can easily refresh memories and use it as a visual reference. UAVs and CRC are photography capture tools that can assist in reexamining visual data. Besides, their captured photos can be used for 3D object reconstruction via photogrammetry process. In order to achieve delicate 3D information capture, scanner-based devices, such as TLS, MHS, and TS, are proven to have higher accuracy (El-Din Fawzy, 2019). TS with the feature of attaining accurate geo-reference, is usually utilized as a ground control reference that supports the scan alignment of TLS or drone photogrammetry alignment, further calibrating point cloud data. Except for conventional camera-based drones, currently, LiDAR sensor-mounted drones are in practice for archaeological excavation sites (Adamopoulos et al., 2023), city models (Setyawan et al., 2022), etc. With ideal weather conditions, the LiDAR drone can capture accurate and reliable information, but it also enormously increases the cost. Considering current UAVs market development, UAVs are still largely camera-based drones which can take photographs and assist with photogrammetry techniques to generate point cloud. Even photogrammetry techniques generally have lower tolerance than scanner-based devices, it provides great flexibility and affordable cost. In addition, UAVs’ data can be upgraded by real-time kinematic (RTK) techniques and achieve much higher and centimeter-level UAVs positioning accuracy in real-time (Czyża et al., 2023), which further support photogrammetry accuracy. Different reality capture tools should be selected or integrated to tackle the aims of project.

Table 1

Table 1. Correlation between project goals and suitable reality capture tools.

Besides different project goals, site limitations can also restrict reality capture tools. Table 2 presents how site constraints, such as interior and exterior conditions, lighting, and accessibility, affect the feasibility of using each tool. Conducting a preliminary on-site investigation is necessary for identifying data capture challenges such as physical access, vegetation, or lighting conditions that can dramatically impact the quality of the data capture process (Fobiri et al., 2022; Willkens, 2022). Most of the tools can be used for interior capture. Scanner-based devices using laser beams can be conducted in a dark environment and still extract good-quality 3D information. Camera-based devices usually need external light sources to collect texture information. Restricted by indoor space, flying a drone in interior space can be a challenge. UAVs for indoor utilization are being studied so far; for example, micro drones have been tested for interior mapping and demonstrated sufficient quality for exploration purposes (Karam et al., 2022). All those tools can be used for exterior capture but might be limited by location and environmental features. UAVs are usually used for hard-access area capture and are suitable for congested areas. MHS has relatively smaller sizes, which usually can be used to scan narrow areas as well but limited to human hand-reachable areas. Scanner-based devices can resist dim light situations when collecting object 3D information, but they often get impacted on collecting material texture and color information on dim environment. Camera-based devices are easily affected by lighting situations. The newest version of Matterport Pro 3 has a LiDAR sensor embedded, which makes it more capable of diverse lighting situations than the last version of Matterport. Previous research (Liu et al., 2023) revealed that a Matterport-captured point cloud would have slightly lower quality and accuracy than a TLS-captured, and Matterport also collected more noise data. However, it is still a sufficient method for assisting with HBIM model-making. Moreover, the allowable time on site is a critical factor to consider. For instance, scanning tourist areas requires avoiding crowds in the daytime, which needs a dedicated scanning plan with proper reality capture tools. Drones and Matterport are deemed as fast scanning devices since drones have flight path and automatic execution features on mapping tasks, and Matterport can do quick 360-degree photography captures. TLS can also achieve quick capture with low-quality and low-resolution parameters setting, while its essential advantage is high-quality and high-resolution capture which can take much longer time to scan. In general, evaluating the site situation is an essential step before comprising a detailed scanning plan. Table 1 and 2 allows users to weigh trade-offs among precision, accessibility, environmental conditions, resource availability, etc. By adjusting the relative weight of each constraint, users can replicate or modify the workflow across diverse built heritage scenarios, from archaeological sites to large-scale urban regions.

Table 2

Table 2. Site condition limitations and applicability of reality capture tools.

Resource availability varies for different projects and stakeholders. Conducting a reality capture task can be burdensome on financial cost, including costs on equipment, processing software, computer station, and reality capture experts. Setting up an achievable goal with available funding resources is the key to the project’s success. In most scenarios, singular reality capture device to finish the whole project scanning work is challenging. Different equipment and software can be combined to finish one project together with good coverage. For example, TLS collects data from the ground level, while UAV gathers data from an aerial view as supplementary information. TLS requires the scan processing software to align all scan positions into one coordinate system, while aerial photos need photogrammetry software to process images and reconstruct the object’s 3D information. As the article mentioned, each reality capture tool has pros and cons and diverse date quality parameters. Usually, the higher resolution data would require more expensive tools and sometimes cost more in terms of scanning time and processing stations. The tool selection should be based on project goals and resource availability to determine.

The project goal, site limitation, and resource availability are essential factors in determining the optimal reality capture plan. At the initial stage of the project, stakeholders meeting is critical to enumerate all requirements as three essential factors can be impacted by each other (Figure 1). In addition, the site situation should be informed in order to come up with a scanning strategy that ensures good coverage rate of scan. When the situation allows, reality capture experts should inspect the site before executing the task to identify any unapparent challenges from the expert’s perspective. To guide practitioners make decisions on capture strategy, Figure 2 introduces a transferable decision flowchart with four core factors in sequence that balances project goals, site constraints, and resource availability, as well as devices capabilities in Tables 1 and 2. Rather than selecting devices solely based on technical specifications, this framework evaluates four core factors in sequence: (1) Available resources and devices; (2) Heritage significance of building component; (3) Permissible tolerance and required LoD; (4) Practical constraints such as budget, time, and site accessibility. The practitioners can evaluate these four core factors in sequence when facing competing constraints. This yields a structured and repeatable method for prioritizing devices under real case conditions. For example, when heritage significance is high and the project requires geometric accuracy with low tolerance, the practitioners should prioritize TLS capture for building elements and area with higher heritage value, even under strict time or budget constraints. The rest of building space and elements capture can shift toward Matterport scanning to achieve rapid capture or shift toward UAVs photogrammetry to overcome labor safety and accessibility constraints. When the project can only get Matterport to use and heritage significance is high, applying high quality capture mode of Matterport on major important area while applying standard quality capture in transition area between major essential space. For adaptive reuse project, prioritizing TLS capture for heritage significance elements and desirable preserving area. Conversely, the space, which is going to be altered heavily and requires LoD is in medium level, can use Matterport scanning served as visual reference record and fast survey device without compromising rehabilitation planning. These scenarios demonstrate how the decision framework can be adapted with competing constraints and provide a transferable methodology that practitioners can apply to varied built heritage project contexts.

Figure 1

Figure 1. Criteria of reality capture plan.

Figure 2

Figure 2. Decision flowchart with four core factors in sequence.

3 Results–case study

3.1 Building background

The Odd Fellows Building, located in the Sweet Auburn Historic District of Atlanta, Georgia, is selected as a case study to demonstrate and verify the HBIM reality capture plan methodology. As an established leader in the African-American community, Booker T. Washington (1856–1915), dedicated the Odd Fellows Building when it was dedicated in 1912. Initially, the building had six stories above ground and one underground floor. Over time, a penthouse floor was added. The building façade was made of red brick except for the first floor, which was made of stone. Amid the Jim Crow era, it was a vital landmark for the Atlanta community, providing office and retail space for Black professionals and Black-owned businesses, while also hosting myriad social activities and cooperative events. With its legacy in supporting the African-American community, Odd Fellows Building was added to the National Register of Historic Places in 1975. However, due to a lack of maintenance and tornado damage in 2008, the building was damaged, and its state severely hindered its legacy and ability to support community needs. Fortunately, with the new adaptive reuse proposal led by a non-profit, Georgia Works, the building will transform into a multi-use complex with office space, discounted retail space, and a transitional housing program to reduce recidivism for Georgia Works program participants. Through these rehabilitations, the building will once again serve as a significant mutual aid function, while conserving the distinctive brick and terracotta of this architectural landmark.

3.2 Reality capture plan

Before handling reality capture tasks and starting digital modeling, examining the project’s circumstances is needed as following the HBIM reality capture plan methodology (Figure 3). The overall goal of adaptive reuse is evident. However, when funding is limited at the initial stage, the project steward aims to obtain an initial digital model to support grant applications, indicating that the budget for scanning and modeling is restricted. Having only a 1-day scanning period and a 1-week modeling budget for this eight-story building makes the selection of suitable reality capture tools challenging. Besides, due to the site limitation of inaccessibility on the North and West sides and rooftop of penthouse, multi-devices have to be integrated to collect complete information. Both TLS and Matterport devices need physically move and set up on every scan location which limits its capture position within a physically reachable area. UAVs can go beyond physical limitation within open air space but becomes risk in congested or indoor space. From labor safety consideration, UAVs can reach out to high elevation and hazard area where historic buildings often have this type of concerns for unstable structure (Treccani et al., 2024). Combining those considerations on project goals, site limitation, and resources availability, the steward determined to collaborate with the institution, which is more accessible and affordable solution and set the desired LoD for the model is 300. Collaborating with a university research institution provided both economic and technical advantages. Unlike professional survey firms that charge commercial rates for fieldwork and modeling, academic collaborations operate under research or educational frameworks that leverage institutional equipment, software licenses, and trained personnel. This project collaborated with the School of Architecture, Georgia Institute of Technology, which possesses advanced reality capture tools, including Faro TLS, Matterport Pro 3, and UAVs. Moreover, faculties and graduate researchers possess pioneer experience with heritage documentation methodologies, data registration, and HBIM workflows, ensuring high-quality outcomes despite limited budgets. This arrangement not only reduces financial expenses but also fosters knowledge exchange between practitioners and the academic team, aligning with the broader educational mission of training future professionals.

Figure 3

Figure 3. (Left) considerations for Odd Fellows Building; (right) site access limitation.

After a comprehensive evaluation, the scanning plan and strategy of using different tools are developed to achieve the general goal (Figure 4). With short time window circumstances, although TLS is deemed the most accurate survey tool (Lachat et al., 2017), it can only be utilized for a small portion since it will consume much longer on-site time than others. Faro Focus S350, as one of the high-end TLS, is applied to capture mainly exterior data and the first-floor area, whose point cloud serves as a baseline for other point clouds to align. Matterport Pro 3 embedded LiDAR sensor is a suitable tool for executing interior scans, as it can operate fast and provide virtual tour as supplementary visual data. Visual data can assist BIM modelers for digital modeling and serve as a photography record as time goes by. The Matterport is set to scan all eight floors, including the basement and rooftop penthouse. Due to the inaccessibility of the North and West facades, UAV, one of the great mobile robots, can easily reach out to areas that human labor cannot access. DJI Mavic 2 takes aerial images to compensate for deficient capture by other tools.

Figure 4

Figure 4. (Left) Odd Fellows Building scan plan strategy; (right) color-coded device coverage diagram showing scanning contributions from faro focus S350 (red), matterport pro 3 (blue), and DJI mavic 2 (yellow).

3.3 Data capture-process-modeling workflow

The overall workflow from capture to processing and modeling is summarized in Figure 5. Aligning the digital documentation strategy with the building’s heritage values, TLS was used to document the lobby corridor and two essential facades (South and East) retained original materials and design intent from the building’s period of significance, as defined in its National Register listing. A total of 14 scan positions were captured by TLS. Matterport scanning was utilized for the renovated upper floors and other non-original areas, including interior space of basement level and upper seven stories where rapid capture and visualization were sufficient for rehabilitation planning but detailed heritage documentation was less critical. Total 205 scan positions were captured by Matterport Pro 3. UAV serve as supplementary reality capture device that collect aerial photos for inaccessibility places of high elevation area and North and West facades. A total of 483 aerial s were collected by DJI Mavic 2. Matterport technology has cloud processing systems that do not require external processing stations. After a Matterport capture, the captured scans can automatically be uploaded to a cloud computing system to generate one integrated virtual tour model directly in their platform. The scan setting, registration accuracy and point density of each capture devices were summarized in Table 3. The TLS dataset achieved a mean point error of 2.6 mm, confirming its role as the prioritized geometric baseline. Since Matterport’s point cloud are pre-aligned in a proprietary coordinate system without ground control referencing. Aligning these data required iterative transformations in Faro Scene and the use of shared reference features such as floor planes and doorframes to reduce geometric drift. Scanning time of different devices further highlights the efficiency of utilizing the combined strategy to balance accuracy and time limitation.

Figure 5

Figure 5. Date capture-process-modeling workflow.

Table 3

Table 3. Quantitative summary of each device’s capture setting and results.

Toward the project goal of having a point cloud for assisting HBIM model making, the virtual tour data is converted as an E57 file, a vendor-neutral format containing 3D point cloud information. After registering in the Faro Scene, the Matterport scanned data serves as an interior point cloud. The data collected from Faro Focus S350 is registered and integrated by Faro Scene to produce a point cloud that will be used as a baseline reference. Aerial images collected by DJI Mavic 2 are processed by photogrammetry techniques in Agisoft Metashape software, whose point cloud is utilized as supplementary information to incomplete scan area by TLS and Matterport. The three separate point clouds have their individual coordinate systems from TLS, Matterport, and UAV (Figures 6a–c), which can cause disorder in the BIM modeling procedure if they have not been aligned. Faro Scene can assist in finding overlapping areas and combining different point clouds. In the HBIM environment, these multi-source point clouds require careful preprocessing to ensure geometric and semantic consistency. As the point cloud from TLS was deemed the accurate baseline, the interior’s point cloud from Matterport was majorly aligned with points in first floor area from TLS, while the point cloud from UAV capture was mainly aligned with points in TLS’s exterior area.

Figure 6

Figure 6. Integrated point cloud datasets captured by TLS, matterport, and UAV. (a) Point cloud from TLS. Capture area: South and East facades and ground floor lobby and corridor. Registration accuracy: 2.6 mm mean point error. (b) Point Cloud from Matterport Pro 3. Capture area: interior floors above ground and basement level. Depth points: ∼0.8 M per position. (c) Point Cloud from UAV photogrammetry. Capture area: inaccessible roof, North and West facades, and high elevation facades area at South and East. RMS reprojection error: 0.311763 (1.21737 pix). (d) Aligned Point Clouds. The alignment among datasets is achieved through overlapping ground floor lobby and corridor, and façade features. Manual marked reference points are used to assist software alignment.

Once the multi-source point clouds aligned (Figure 6d), the integrated dataset was imported into Autodesk ReCap Pro for final filtering and down sampling to optimize file size and maintain manageable density for HBIM modeling. The registered point cloud was then linked into Autodesk Revit, where model development followed a hierarchical, element-based approach consistent with HBIM best practices. Major components, including structural columns, floor slabs, walls, windows, and staircases, were modeled as Revit families using the Architecture and Structure toolsets. To maintain geometric fidelity, elements were traced directly over the point cloud. Material properties were documented through parameterized attributes and labeled according to the LoD 300 definition, which ensures accurate geometric representation and basic non-geometric information (Figure 7).

Figure 7

Figure 7. (Left) HBIM model with LoD 300; (right) HBIM model with point clouds on.

4 Discussion and conclusion

4.1 Discussion

The case study of the Odd Fellows Building demonstrates the practical advantages and constraints of implementing a multi-device HBIM workflow in adaptive reuse projects. By integrating TLS, Matterport, and UAV technologies, the project achieved comprehensive coverage of interior and exterior conditions under severe time and budget constraints. This hybrid strategy illustrates a valuable trade-off: TLS provided highly accurate baseline data, Matterport delivered rapid interior scanning with the added benefit of immersive visualization, and UAVs captured inaccessible areas that would otherwise be excluded from conventional surveys. The resulting dataset, once registered and unified, offered stakeholders both the precision required for design decisions and the accessibility needed for communication and funding efforts.

These findings underscore the potential of multi-device workflows to address persistent challenges in heritage documentation. Previous studies have often emphasized either the precision of TLS or the accessibility of photogrammetry and low-cost scanning technologies. This case instead demonstrates the value of strategically combining devices to balance accuracy, speed, and resource limitations. The Atlanta case study successfully achieved comprehensive coverage of both exterior and interior spaces within the allotted 1-day scanning and 1-week modeling timeframe, delivering a unified LoD 300 HBIM model. Despite limited resources, this outcome met all project goals for documentation completeness, LoD requirement, and assisting stakeholder communication on initial rehabilitation proposal. The workflow also highlights the evolving role of HBIM as a bridge toward DT applications: while the resulting model does not achieve real-time monitoring, it establishes the foundational infrastructure necessary for future data integration and immersive exploration.

Nevertheless, the study also points to important limitations. The reliance on multiple devices requires expertise in data registration and integration, which may present barriers for projects without institutional or technical support. The integration of heterogeneous datasets also revealed significant HBIM challenges. Aligning point clouds from TLS, UAV, and Matterport required a robust data management protocol to avoid geometric distortion and ensure interoperability across software platforms, such as Faro Scene, Matterport, Agisoft Metashape, and Autodesk Revit. The accuracy assessment of aligned point cloud was primarily based on visual alignment and sample measurements. A comprehensive quantitative validation using standardized error metrics and control targets would strengthen its reliability. Another major constraint involves file size and computing resource management. Integrating multiple high density point clouds often produce large datasets, requiring progressive down sampling and segmentation to maintain Revit performance without compromising accuracy. This challenge highlights the importance of establishing standardized data resolution thresholds for HBIM projects. Besides, the HBIM model was developed to LoD 300, which is suitable for design coordination at this stage but not for detailed conservation analysis. Historically significant elements such as decorative brickwork and terra cotta would require further detail modeling and semantic data enrichment to assist preservation work. Furthermore, the workflow relies on commercial platforms such as Matterport’s hosting service, raising questions about licensing costs, data interoperability, and long-term accessibility in heritage contexts. Future research should pursue higher LoD and semantically enriched models, which can support DT integration to enable dynamic monitoring and preventive maintenance applications. By incorporating the HBIM model with sensor data, environmental monitoring, and maintenance logs, users can transition from static representation to a dynamic performance model capable of real-time tracking and predictive analysis, advancing from documentation toward preventive conservation and life-cycle management of historic assets.

4.2 Conclusion

This paper presented a multi-device HBIM workflow tested through the adaptive reuse project of the Odd Fellows Building in Atlanta. By leveraging TLS, Matterport, and UAV technologies, the project developed a rapid, resource-conscious, and semantically rich HBIM model at LoD 300. The workflow demonstrates how hybrid capture strategies can overcome site access limitations, accelerate documentation, and support both stakeholder communication and renovation planning.

Importantly, decisions about which technologies to use were guided by considerations of authenticity and integrity, aligning the digital documentation strategy with the building’s heritage values. Since the interior had undergone extensive alterations, only the lobby corridor and two primary facades (south and east) retained original materials and design intent from the building’s period of significance, as defined in its National Register listing. These areas were documented with TLS to ensure the highest geometric accuracy and tolerance. In contrast, Matterport scanning was prioritized for the renovated upper floors and other non-original areas, where rapid capture and visualization were sufficient for rehabilitation planning, but detailed heritage documentation was less critical. This selective approach exemplifies how HBIM workflows can integrate both technical precision and preservation ethics, balancing the documentation of historically significant elements with pragmatic resource allocation.

Beyond the immediate project outcomes, the study contributes to the broader discourse on digital transformation in construction by positioning HBIM as a scalable framework that integrates diverse capture technologies while remaining sensitive to the cultural and material authenticity of historic buildings. The Odd Fellows case underscores the importance of balancing accuracy, efficiency, and significance-based decision-making in heritage contexts where time, funding, and access are constrained.

Future research should focus on refining interoperability between commercial and open-source platforms, exploring automation in point cloud registration, and testing the scalability of significance-driven multi-device workflows across different heritage typologies. By advancing such approaches, HBIM can more fully realize its role in preserving cultural heritage with integrity while supporting the digital transformation of the construction sector.

Data availability statement

The datasets presented in this article are not readily available because the data supporting this study’s findings are available upon request from the corresponding authors. Due to the sensitive nature of the data and their file size, access to the data will be granted on a case-by-case basis and may require a data use agreement to be signed. Requests to access the datasets should be directed to DW, ZGFuaWVsbGUud2lsbGtlbnNAZGVzaWduLmdhdGVjaC5lZHU=.

Author contributions

BL: Methodology, Visualization, Data curation, Software, Writing – original draft, Conceptualization, Writing – review and editing. DW: Data curation, Conceptualization, Supervision, Methodology, Project administration, Writing – review and editing, Software.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Adamopoulos, E., Papadopoulou, E. E., Mpia, M., Deligianni, E. O., Papadopoulou, G., Athanasoulis, D., et al. (2023). 3D survey and monitoring of ongoing archaeological excavations via terrestrial and drone lidar. ISPRS Ann. Photogrammetry Remote Sens. Spatial Inf. Sci. X-M-1-2023, 3–10. doi:10.5194/isprs-annals-x-m-1-2023-3-2023

CrossRef Full Text | Google Scholar

Alshawabkeh, Y., and Baik, A. (2023). Integration of photogrammetry and laser scanning for enhancing scan-to-HBIM modeling of Al Ula heritage site. Herit. Sci. 11 (1), 147. doi:10.1186/s40494-023-00997-2

CrossRef Full Text | Google Scholar

Cassar, A., Baker, D., and Ioannides, M. (2025). From digital twin to memory twin: a holistic framework for cultural heritage documentation, interpretation, and adaptive reuse. Int. Arch. Photogrammetry Remote Sens. Spatial Inf. Sci. 203–209. doi:10.5194/isprs-archives-xlviii-m-9-2025-203-2025

CrossRef Full Text | Google Scholar

Czyża, S., Szuniewicz, K., Kowalczyk, K., Dumalski, A., Ogrodniczak, M., and Zieleniewicz, Ł. (2023). Assessment of accuracy in unmanned aerial vehicle (UAV) pose estimation with the real-time kinematic (RTK) method on the example of DJI matrice 300 RTK. Sensors 23 (4), 2092. doi:10.3390/s23042092

PubMed Abstract | CrossRef Full Text | Google Scholar

Dore, C., and Murphy, M. (2017). Current state of the art historic building information modelling. Int. Arch. Photogrammetry Remote Sens. Spatial Inf. Sci. 185–192. doi:10.5194/isprs-archives-xlii-2-w5-185-2017

CrossRef Full Text | Google Scholar

El-Din Fawzy, H. (2019). 3D laser scanning and close-range photogrammetry for buildings documentation: a hybrid technique towards a better accuracy. Alexandria Eng. J. 58 (4), 1191–1204. doi:10.1016/j.aej.2019.10.003

CrossRef Full Text | Google Scholar

Fobiri, G., Musonda, I., and Muleya, F. (2022). Reality capture in construction project management: a review of opportunities and challenges. Buildings 12 (9), 1381. doi:10.3390/buildings12091381

CrossRef Full Text | Google Scholar

Giuliani, F., Gaglio, F., Martino, M., and De Falco, A. (2024). A HBIM pipeline for the conservation of large-scale architectural heritage: the city Walls of Pisa. Herit. Sci. 12 (1), 35. doi:10.1186/s40494-024-01141-4

CrossRef Full Text | Google Scholar

Inzerillo, L., Pisciotta, A., Acuto, F., Mantalovas, K., and Di Mino, G. (2025). Beyond traditional H-BIM: AI-powered modeling for heritage bridges. Int. Arch. Photogrammetry Remote Sens. Spatial Inf. Sci. 615–622. doi:10.5194/isprs-archives-xlviii-m-9-2025-615-2025

CrossRef Full Text | Google Scholar

Karam, S., Nex, F., Karlsson, O., Rydell, J., Bilock, E., Tulldahl, M., et al. (2022). Micro and macro quadcopter drones for indoor mapping to support disaster management. ISPRS Ann. Photogrammetry Remote Sensing Spatial Inform. Sci. V-1-2022, 203–210. doi:10.5194/isprs-annals-v-1-2022-203-2022

CrossRef Full Text | Google Scholar

Khalil, A., and Stravoravdis, S. (2022). Challenges of digital building data usage with a focus on the digital documentation of heritage buildings—results from an online survey. Heritage 5 (4), 3220–3259. doi:10.3390/heritage5040166

CrossRef Full Text | Google Scholar

Khan, M., Khan, M., Bughio, M., Talpur, B., Kim, I., and Seo, J. (2022). An integrated HBIM framework for the management of heritage buildings. Buildings 12 (7), 964. doi:10.3390/buildings12070964

CrossRef Full Text | Google Scholar

Lachat, E., Landes, T., and Grussenmeyer, P. (2017). Investigation of a combined surveying and scanning device: the trimble SX10 scanning total station. Sensors 17 (4), 730. doi:10.3390/s17040730

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, J., Willkens, D., López, C., Cortés-Meseguer, L., García-Valldecabres, J. L., Escudero, P. A., et al. (2023). Comparative analysis of point clouds acquired from a TLS survey and a 3d virtual tour for Hbim development. Int. Arch. Photogrammetry Remote Sens. Spatial Inf. Sci. 959–968. doi:10.5194/isprs-archives-xlviii-m-2-2023-959-2023

CrossRef Full Text | Google Scholar

Lucchi, E. (2023). Digital twins for the automation of the heritage construction sector. Automation Constr. 156, 105073. doi:10.1016/j.autcon.2023.105073

CrossRef Full Text | Google Scholar

Niccolucci, F., and Felicetti, A. (2024). Digital Twin sensors in cultural heritage ontology applications. Sensors 24 (12), 3978. doi:10.3390/s24123978

PubMed Abstract | CrossRef Full Text | Google Scholar

Parente, M., Bruno, N., and Ottoni, F. (2025). HBIM and information management for knowledge and conservation of architectural heritage: a review. Heritage 8, 306. doi:10.3390/heritage8080306

CrossRef Full Text | Google Scholar

Rocha, G., Mateus, L., and Fernández, J. (2020). A Scan-to-BIM methodology applied to heritage buildings. Heritage 3, 47–67. doi:10.3390/heritage3010004

CrossRef Full Text | Google Scholar

Setyawan, A. A., Taftazani, M. I., Bahri, S., Noviana, E. D., and Faridatunnisa, M. (2022). Drone LiDAR application for 3D city model. J. Appl. Geospatial Inf. 6 (1), 572–576. doi:10.30871/jagi.v6i1.3462

CrossRef Full Text | Google Scholar

Treccani, D., Adami, A., Fregonese, L., Treccani, D., Adami, A., and Fregonese, L. (2024). Drones and real-time kinematic base station integration for documenting inaccessible ruins: a case study approach. Drones 8 (6), 268. doi:10.3390/drones8060268

CrossRef Full Text | Google Scholar

Willkens, D. S. (2022). The craft and care of reality capture. Build. Landscapes J. Vernac. Archit. Forum 29 (2), 2–14. doi:10.1353/bdl.2022.0009

CrossRef Full Text | Google Scholar

Yang, X., Grussenmeyer, P., Koehl, M., Macher, H., Murtiyoso, A., and Landes, T. (2020). Review of built heritage modelling: integration of HBIM and other information techniques. J. Cult. Herit. 46, 350–360. doi:10.1016/j.culher.2020.05.008

CrossRef Full Text | Google Scholar

Yu, Y., Raed, A. A., Peng, Y., Pottgiesser, U., Verbree, E., and Van Oosterom, P. (2025). How digital technologies have been applied for architectural heritage risk management: a systemic literature review from 2014 to 2024. Npj Herit. Sci. 13 (1), 45. doi:10.1038/s40494-025-01558-5

CrossRef Full Text | Google Scholar

Zhang, Z., and Zou, Y. (2022). Research hotspots and trends in heritage building information modeling: a review based on CiteSpace analysis. Humanit. Soc. Sci. Commun. 9 (1), 394. doi:10.1057/s41599-022-01414-y

CrossRef Full Text | Google Scholar