Integrating thermal point clouds into BIM-GIS environments: workflow proposal for multi-layer digital twins

Introduction: The integration of Thermal Point Clouds (TPCs) into professional Building Information Models (BIM) and Geographic Information System (GIS) workflows is currently hampered by a lack of established methodologies and significant format interoperability challenges. This study addresses this methodological gap by developing and testing integration processes for both BIM and GIS platforms.

Methods: Two proofs of concept were developed using data from a commercial scanner at the New York University Abu Dhabi (NYUAD) campus. For BIM integration, a process was designed to generate open Industry Foundation Classes (IFC) files where average surface temperatures are embedded as native properties of architectural elements. For GIS integration, thermal data was assigned as custom attributes to a manually generated 3D geometric reference model, establishing the preliminary steps for a dedicated thermal-GIS workflow.

Results: The methodologies were successfully validated through visualization in ArcGIS Pro and ACCA Software GeoTwin. The results demonstrate a tangible path to overcoming current format limitations, enabling the creation of multi-layer thermal digital twins.

Discussion: This approach makes complex thermal data more accessible to Architecture, Engineering, Construction, and Operation (AECO) professionals. By providing a structured workflow for interoperability, the study facilitates improved building management and more accurate energy analysis through the use of integrated thermal digital models.

1 Introduction

The Architecture, Engineering, Construction, and Operations (AECO) industry is in a constant search for innovative solutions to understand the current state of buildings and infrastructure, monitor different variables that can improve management, seek greater energy efficiency in the built environment, and move towards zero-energy buildings and infrastructure. In this context, digital twins are already a reality in the industry, employing methodologies such as Building Information Models (BIM) and Heritage Building Information Models (H-BIM) to create information models that can be enriched with a multitude of data. Over the last decade (2015–2025), numerous studies have also emerged proposing the contextualization of these building information models in broader urban settings, where georeferenced data can be included using GIS systems. For example, López-González and García-Valldecabres (2023), explored and determined the possibilities of interoperability between H-BIM and GIS to synchronize and manage information on heritage architecture, sustainable conservation, and territorial and cultural tourism planning, developing a novel technological system that integrates H-BIM and GIS into a single tool for cultural development and heritage preservation. A 3D-GIS prototype is created for visitor management and preventive maintenance of heritage buildings and centers using the ArcGIS Pro platform, controlling problems caused by an excessive number of visitors in the conservation of a monument through sensors (CO2, humidity, temperature) and connecting this data with the number of visitors. Bottaccioli et al. (2017) presented a cloud-based software architecture for managing and simulating the energy performance of buildings, integrating heterogeneous data (BIM, IoT, GIS, weather services) to allow for (near) real-time visualization of energy consumption, building performance evaluation through modeling and simulation, and the creation of GIS energy maps. In this way, models are created that are enriched with information from all kinds of sensors to contribute to the monitoring of buildings and infrastructure, which can also help in the analysis and study of, among other aspects, energy efficiency. Back et al. (2023) also used a GIS model to improve urban heat and thermal comfort assessment, considering the vertical and horizontal variability of wind speed and air temperature patterns through computational fluid dynamics (CFD) simulations in an urban case study model in Innsbruck, Austria. Mutani et al. (2016) compared two GIS-based methodologies for the spatial characterization of the residential built environment at an urban scale in terms of building distribution and energy consumption for heating. Additionally, they proposed a third method for non-residential buildings, with the aim of supporting local energy planning and creating urban energy maps.

Despite their clear potential, the integration of thermal data in AECO remains an under-researched area. Currently, most projects focus on pairing 2D thermal images with geometric models, which limits the management of temperature data for applications such as energy simulations or thermal comfort studies. There are different methodologies for collecting thermal data for their inclusion in urban or building digital models, such as the one proposed by Lin et al. (2019), which fuses data from terrestrial thermal images with 3D point clouds to map thermal attributes on building facades, focusing on a registration between the thermal images and the point clouds. Thermal and RGB point clouds are created from their respective images using Structure from Motion (SfM) tools. In this case, the RGB point cloud is used as a 3D reference. In Hou et al. (2021), the performance of the fusion of exterior thermal data from facades is evaluated, considering how different Unmanned Aerial Vehicle (UAV) flight configurations, such as altitude, camera angle, flight path design, and construction typologies, affect it. This study seeks to provide suggestions for the collection and better processing of this type of data for the diagnosis of pathologies and failures in energy efficiency.

Within these types of digital models, the inclusion of thermal information has gained significant relevance in the last 5 years (2020–2025), allowing for the collection of the three-dimensional geometry of a building or space along with temperature data in continuous monitoring campaigns, and the correspondence between these points and temperatures, as developed in Adán et al. (2020), and the work carried out by Ramón-Constantí et al. (2022a). This information is initially presented as an unstructured TPC, which, once processed, can be transformed into a Proprietary Thermal Model (TPM). TPCs offer valuable applications in the AECO industry by providing a detailed and accurate representation of the distribution of surface temperatures on the interior and exterior of buildings, which makes them useful tools for understanding the thermal behavior of constructive elements and monitoring thermal changes over time. Additionally, they can be segmented and tagged for its use in BIM environments, containing specific information about the thermophysical characteristics of materials, facilitating maintenance tasks such as the identification of singular regions like thermal bridges, as well as serving as input data for performing energy simulations, as developed in Ramón-Constantí et al. (2024), where an experimental methodology is developed to work with this type of thermal model and calibrate building energy simulations using CFD. The integration of thermal data with other aspects of the architectural project also promotes multidisciplinary collaboration and more efficient decision-making regarding the management and maintenance of buildings.

Despite their clear potential, the use and applicability of building TPCs in professional software nowadays remains limited. This is largely due to significant challenges related to interoperability and format compatibility, which hinder the incorporation of thermal data and the subsequent management of proprietary thermal models in BIM and GIS environments. By laying the groundwork for a broader use of thermal models in the management and conservation of buildings, this work seeks to establish the path toward the application of thermal digital twins in the construction sector through a use case using a BIM-GIS integration.

The article is organized as follows: Section 1 is an introduction to the topic that discusses the current status and limitations in the AECO industry regarding thermal models and the poor state of the art regarding the integration of TPC into BIM and GIS; Section 2 presents the objectives, general methodology and contributions of the study; Section 3 describes the case study; Section 4 describes the use and integration of TPCs in an open-BIM viewer; Section 5 describes the integration of TPCs in a GIS platform; Section 6 discusses the limitations of the methodologies; Section 7 shows the conclusions of the study, and Section 8 proposes future works regarding the acquisition, implementation and use of TPCs in BIM-GIS scenarios.

1.1 The irruption and potential utility of the as-is thermal models in AECO

There has been a growing interest in systems and methodologies that allow the digitalization of buildings in their current state of the art. The term “as-is” (or “as-built”) refers to a record of the dimensions, components and other aspects of an existing building or infrastructure at a time after its construction has been completed. In this framework, the aim is to optimize a workflow that, until now, has depended mainly on a technician who performs many of the processes manually and handles large volumes of data and information necessary to understand the operation and improve the preventive maintenance of buildings.

Traditionally, 3D digitalization of buildings has focused on capturing their geometry. However, in the last decade, studies and proposals have opened new lines of research aimed at the analysis and study of the energy efficiency of existing buildings, as pointed out by Rahhal et al. (2019). This has led to the first works regarding thermal 3D digitalization of buildings, and as a result, digital twins can now be enriched with this new data. Desogus et al. (2023) developed a BIM-centric workflow for monitoring, visualizing, and assessing thermal comfort in buildings, integrating BIM tools and sensor technology through visual programming using Dynamo visual programming interface (VPI). The objective is to optimize building energy use while achieving good indoor environmental conditions and supporting facility managers by updating the as-built model with efficiency improvements and linking parameters for monitoring. Monitoring data is associated with the “room” object in the proposed BIM case study. In Dlesk et al. (2023) it is proposed to optimize facility management by integrating IoT infrastructure with digital building models, enabling informed decisions using spatially oriented sensor data and creating a digital twin aimed at improving users’ thermal comfort and reducing energy consumption. Thermal 3D digitalization entails collecting the three-dimensional geometry of a building or space, temperature data, and matching points and temperatures. The whole information is originally presented as an unstructured TPC, which can be subsequently processed until obtaining a Thermal Proprietary Model (TPM). In this regard, recent research into the generation of 3D models with thermal information stands out, although its scope is still very limited to the academic field, as concluded in the study by Ramón-Constantí et al. (2022b).

TPCs can have different useful applications in the AECO industry as they provide a detailed and accurate representation of the distribution of surface temperatures in particular areas or elements of a building or facility. Thus, they are tools to understand the thermal behaviour of construction elements, rooms or groups of rooms, as well as for monitoring thermal changes over time, as shown in Adán et al. (2023b), where an automated robotic platform is presented for obtaining thermal maps of interiors at different times through an automatic data acquisition approach. They are also useful for inferring thermal characteristics of buildings, as demonstrated by Hou et al. (2021) which proposed a framework for RGB-thermal data fusion to create district-level thermal maps, evaluating the impact of flight configurations such as camera altitude, angle, flight path design, and building types, and the similar study by Alba et al. (2011) where a 3D acquisition and processing method that integrates Red, Green and Blue (RGB) colour, thermal infrared (IR), and near-infrared (NIR) imagery onto detailed 3D building models is presented. The approach streamlines data fusion for anomaly detection and conservation analysis, using a “bi-camera” system, photogrammetry, and terrestrial laser scanning.

One of the advantages of these models is that they can be segmented and tagged for use in BIM environments, containing specific information related to other attributes of the material thermophysical characteristics. This allows working with local or global temperature data that can be used for various maintenance tasks, from the identification of heat loss regions to the evaluation of energy performance over time of residential buildings. Furthermore, integrating thermal data with other aspects of the architectural project facilitates collaboration between multidisciplinary teams, thus enabling informed and more efficient decision-making. Professionals can also assess the thermal impact of different designs, construction and retrofitting strategies and compare it with historical records and surface temperature readings taken in real time.

Another highlight is the scalability of these models. Thermal models can be enriched by adding new measurements or data coming from local sensor networks or extending the scope of study to other nearby buildings. This means that the models can be adapted to address specific challenges at different stages of the operation phase and long-term maintenance, even if the area of the building to be analysed is expanded.

1.2 Integration problems

There is a significant lack of research regarding the integration of data from different thermal data sources in AEC. Currently, the majority of the projects and research are mainly focused on matching thermal IRF images to geometric models of the facility or building, which does not really allow the construction professionals to manage quantitative temperature data for further applications, such as energy simulations or thermal comfort studies, what it is possible using TPCs, and, if any, the subsequently obtained Thermal Proprietary Models (TPM). A TPM is a basic three-dimensional model generated by programmers after processing the original TPCs of a scene. This model incorporates the surface temperatures of key architectural elements of a building, such as floors, ceilings, walls, columns, doors, and windows, into the geometry, as described by Adán et al. (2023c).

However, the use of TPMs in the AEC industry presents a series of adaptation problems to standard formats that are still little addressed today, especially regarding the variety and inter-compatibility between them. Moreover, adapting these thermal data to standard formats entails a high technical complexity that does not facilitate their widespread use. Therefore, the creation of thermal models in formats that allow their inclusion in digital twins of buildings would lead to a significant advance in applications related to building management and architecture in their operational phase.

Although there are standards, such as IDF or gbXML and others, the lack of a universally accepted format that can include thermal data from buildings makes it difficult to transfer thermal information between different software tools and platforms, both open source and commercial software. This leads to the need to make specific adaptations to the data structures of these models for each software, which increases the complexity and time required to integrate thermal models into existing workflows. In addition, these modifications are usually made using tools and processes that involve specific programming knowledge, so it is not a simple process and has to be carried out by specialists. The most recent references deal with the issue of interoperability between formats in an energy simulation context, as in Bracht et al. (2021), where they develop using Autodesk Revit a gbXML-based schema, but at no point do they work with thermal models as the proposed work is focused in the early design stages. In Kamel and Memari (2019), this topic is also reviewed, but from a BIM-to-BEM point of view, where it is concluded that there are unresolved aspects such as data loss between formats, the improvement of different standards and formats, the use of intermediate tools and the requirement of manual processes to avoid data loss between two tools during data exchange.

Furthermore, the integration of real temperature data into BIM may require adding specific attributes or tags to represent the thermal information natively in each software through manual processes. This is because many BIM software programs are not designed to natively visualize thermal data, such as through color-coded representations. Although there are recent studies that show the integration of IR images in different BIM software on 3D geometric models, such as in Adamopoulos et al. (2021), no studies have been found that show how to integrate TPCs.

In summary, the integration of dense thermal information in the AEC sector currently faces significant challenges in terms of format compatibility and data exchange between open and commercial software, problems related to the capacity of the available software, and the representation and management of thermal data in BIM models.

1.3 TPC in BIM-GIS

Apart from the critical integration problems previously discussed, the use of TPCs has experienced remarkable growth over the last decade (2015–2025), driven by advances in 3D digitalization technologies and the increased accuracy and speed of obtaining data from buildings and infrastructure. In particular, integrating thermal point clouds into BIM and GIS is of great interest, as it can help detect issues related to building maintenance and management in the context of smart cities and the creation of digital twins of the built environment.

Beyond the thermal component, researchers have extensively explored the joint use of BIM and GIS as complementary methodologies that support building analysis and management. Göçer et al. (2016) present an integrated BIM–GIS model for pre-renovation processes, enabling the organization and visualization of relevant data to optimize renovation decisions in economic and environmental terms. Their approach addresses limitations of previous models, such as fragmented information and insufficient analytical tools, thereby improving collaboration between stakeholders. Kamel and Memari (2019) review the application of BIM in energy simulation using tools such as Revit, OpenStudio, and EnergyPlus, identifying persistent issues related to data transfer, information loss, and interoperability constraints across standards such as IFC and gbXML. Oti et al. (2016) propose a framework for integrating building management system (BMS) data into BIM models, demonstrating how operational data can support more sustainable building design and performance evaluation.

As highlighted by Hong-mei et al. (2001), GIS offers strong analytical capabilities for geospatial information. Its integration with BIM has therefore attracted interest in diverse applications, such as traffic noise modelling (Deng et al., 2016), flood damage assessment (Amirebrahimi et al., 2016), and indoor evacuation planning (Isikdag et al., 2013). In García-Valldecabres et al. (2023), a 3D-GIS database integrating HBIM models is implemented to study visitor behavior and mobility in historic centers. Such examples show how the combination of BIM and GIS offers enriched spatial and semantic information suitable for complex building and urban analyses.

Regarding the thermal dimension, incorporating thermal information into BIM–GIS contexts has become increasingly relevant in recent years. Antón and Amaro-Mellado (2021) proposed an open-source GIS-based platform for the visualization and analysis of 3D thermal point clouds of exteriors and urban environments. Previtali et al. (2012) introduced a photogrammetric methodology for mapping infrared images onto 3D models in a GIS environment. Lagüela et al. (2014) presented a workflow for thermal mapping of exterior building models using UAV-mounted thermographic cameras, with comparison against terrestrial laser scanning as reference data. These studies underline the importance of integrating thermal information into BIM-GIS models for applications such as assessing thermal insulation systems, detecting construction defects, or monitoring energy performance. Complementarily, Tomchinskaya and Galanina (2020) address the integration of non-geometric data for infrastructure management in a campus setting, while Niu et al. (2015) develop a BIM–GIS-based system for visualizing building energy data. However, despite these advances, no studies to date describe structured methodologies for integrating full TPC datasets into BIM-GIS systems.

While previous research has focused primarily on mapping 2D infrared imagery onto 3D geometries or on limited thermal-fusion techniques, these approaches do not enable the structured propagation of point-based thermal attributes into standard BIM or GIS formats. Moreover, existing BIM–GIS interoperability studies remain centered on geometric or alphanumeric information and do not address the challenges associated with representing dense, three-dimensional thermal datasets. The workflow proposed in this study addresses this gap by enabling the incorporation of TPC-derived temperature attributes into both an open BIM schema (IFC) and a GIS environment. This dual integration ensures that thermal information can be preserved, queried, and exchanged across platforms, thereby extending current practices beyond IR-image mapping methods and providing a novel, comprehensive pathway for thermal data interoperability in BIM–GIS workflows. Surface temperatures assigned to different elements in the model.

2 Materials and methods

2.1 Objectives of the study

The main objective of this research is to develop a novel methodology for managing TPCs and their associated data within a combined BIM-GIS environment. The goal is to establish a workaround that lays the foundation for the creation of thermal digital twins, designed to enable the various building stakeholders to understand and improve the thermal and energy performance of interior spaces, as well as the comfort of their users. To achieve this goal, the following specific objectives have been defined.

1. To establish a workaround for the integration of TPC into GIS environments. Through a case study, the aim is to create a workaround that allows, for the first time, the direct incorporation of TPC into a geospatial environment, addressing a significant gap in current research.

2. To propose a methodology trying to reduce the need for advanced programming, while acknowledging that certain steps, particularly IFC generation and attribute automation, currently still require use of scripting.

3. To generate a digital replica with thermal properties. The goal is to create a digital twin of the physical structure that incorporates its thermal characteristics. This model should allow for three-dimensional visualization with multiple layers of information (TPC data, sensor readings, etc.), facilitating a detailed analysis of each component.

4. Incorporate time variables for dynamic analysis. The integration of the time dimension into the digital model will be explored, allowing managers to observe and verify the building’s thermal evolution and changes throughout its use cycle.

2.2 Methodology workflow

This research addresses a notable gap in the scientific literature which is the lack of established methodologies for integrating TPCs into GIS environments and the scarcity of studies that do so in BIM environments. While previous works exists, such as that of Ramón-Constantí et al. (2024), which integrates a TPC into a model for building energy simulation (BES) using several CFD simulations of an indoors scenario, direct connection with GIS platforms at the level of three-dimensional thermal data remains an unexplored field. This research seeks to establish a workflow for the efficient generation of a thermal digital twin, a virtual replica that facilitates the integration of geometric and alphanumeric data based on spatial location for its use in building management by different stakeholders.

To develop the methodological framework, existing approaches for integrating BIM and HBIM (Heritage Building Information Modelling) models with GIS were analyzed. For example, the study by Escudero et al. (2024) demonstrates the feasibility of incorporating detailed alphanumeric data for the comprehensive visualization and analysis of 3D models in such environments using an IFC format. Complementarily, the methodology proposed by Albourae et al. (2017) for generating HBIM models offers a relevant workflow which begins with data collection using terrestrial laser scanning (TLS) and photogrammetry; continues with the extraction of a 3D model from the integrated data; and does an element identification within a BIM platform such as Autodesk Revit. A key aspect of their work is the creation of parametric object libraries for distinctive historical elements enabling a detailed study of architectural heritage. They propose integrating the HBIM model into a 3D-GIS to semantically enrich it with non-architectural information, such as materials, historical context, or repair history, overcoming the limitations of traditional 2D-GIS. The methodology used to create the TPC is not the subject of this article since it is already a topic developed in Ramón-Constantí et al. (2022b), Adán et al. (2023a), Adán et al. (2023b). Since there is no established protocol for integrating TPCs into GIS, the methodology proposed in this article is structured as follows:

(1) An exhaustive information search was conducted to determine the state of the art, using sources such as Web of Science and Journal Citation Reports, Google Scholar, and ResearchGate, with a focus on the last decade (2015–2025) and using keywords such as “BIM-GIS,” “thermal point cloud integration,” and “TPC-GIS.”; (2) Data acquisition planning was carried out, which involved familiarization with the available hardware (fixed and mobile sensors) and (3) coordination with the stakeholders involved in the acquisition process, establishing the optimal location for the sensors; (4) The field phase consisted of on-site data collection, followed by (5) downloading and processing the information, generating two products: a non-geolocated TPC model and a geometric model from a manual Scan-to-BIM process in Revit. In parallel, an IFC file was partially-automatically generated from the TPC, containing the average thermal information for each element; (6) The manual 3D model was subsequently imported into a GIS environment, where it was modified to assign the average temperature as a GIS attribute to each element. Finally, (7) it was semantically enriched with open data and information from the available sensors. The result is an integrated model that is imported and public viewed in ArcGIS Online (sceneviewer) and connected to ACCASoftware GeoTwin (for the TPC), serving as the basis for a thermal digital twin that can be used for efficient building management and monitoring which can be used by different building stakeholders in a university or limited urban indoors/outdoors context. The workflow includes both automated and manual processes: scripting is used for vertex extraction, attribute assignment, and IFC generation, whereas segmentation and modelling remain manual. A flowchart of this process is provided in Table 1.

Table 1

Table 1. Diagram of the BIM-to-thermal-GIS methodology, its steps, automatic vs manual process and used software (Partially Automatic means automatic execution of pre-developed scripts).

2.3 Contribution of the paper

This article presents a novel, practical workaround for integrating TPC data into both standard BIM formats and a GIS environment, demonstrating a significant step toward developing thermal digital twins. The contribution lies in formalizing and validating a workflow that has not been described in prior literature, particularly for TPC–GIS integration. This contribution entails a step towards the development of thermal digital twins and aims to establish a first novel workflow for integrating this type of data into BIM-GIS platforms. The novelty of this workflow against prior literature is achieved by incorporating high-resolution, complex thermal data (obtained from large-scale, multi-story glazed façades) into common AECO industry platforms, particularly addressing the current lack of direct, full-feature integration tools between TPC processing software, BIM, and GIS. The methodology for obtaining TPC data of buildings is based on that developed by Ramón-Constantí et al. (2022a), and which was also discussed in Adán et al. (2023a); Adán et al. (2023b), so it is not the object of this article.

The article presents the steps using a case study composed of an interior scene of a building together with a portion of a complex façade. This work has been structured in two parts:

1. Integration of TPCs into open BIM formats: The aim is to import TPCs into the IFC BIM format and, in this way, obtain collaborative and interdisciplinary thermal models in the BIM field.

2. Integration of TPCs into geographic information systems (GIS): The aims are to introduce a TPC into a collaborative environment beyond the building scale, and to share information obtained from different data sources.

3 Case study

In this case study, geometric and thermal data were collected over 10 days in a building located on the New York University Abu Dhabi (NYUAD) Campus. The sensors used were the Leica’s BLK360 laser scanner, a network of ambient temperature and contact sensors, and a set of infrared (IR) images taken with a handheld IR camera. Table 2 specifies the technical characteristics of each of the sensors used.

Table 2

Table 2. Technical characteristics of the sensors and equipment used.

The building studied is located on the west side of the campus and consists of a spacious ground floor with multiple entrances, wide corridors, and areas dedicated to classrooms, laboratories, offices, and open and closed workspaces. The work area includes a ground floor and a mezzanine. The ground floor is mainly dedicated to transit areas, with a large double-height hall that opens onto an interior courtyard built with an ellipsoidal glass curtain wall. Figure 1a shows the setup of one of the scans. Figure 1b shows the interior view of the RGB point cloud inside the building. The upper floor houses circulation areas, open offices, and workspaces that were not accessible during the study. During data collection, the entire interior space was air-conditioned.

Figure 1

Figure 1. (a) Set-up of the laser scanner mounted on the tripod and computer equipment to capture data from a single position; (b) Interior view of the RGB point cloud inside the building.

To perform a complete thermal scan of the scene, the scanner laser sensor was placed in three different positions in the interior of the building: two on the ground floor to capture data from the half curtain wall, one on the landing of the stairs connecting the double height to the office space on the mezzanine and two on the upper floor. Three thermal point clouds (one vertical and two oblique at +45° and −45°) were obtained for each position in order to cover the visible space from the position, resulting in a total of 15 scans. Figures 2a–e provide an example of five of these scans, while Figure 2f illustrates the complete TPC after registering the collected samples. Using the scanner configuration to create TPCs does not allow them to be geolocated; this data is lost during the processing and formation of the thermal cloud.

Figure 2

Figure 2. Sample of thermal point clouds from 5 scanner positions on: the interior ground floor (a,b), the landing of the stairs (c), and the upper floor (d,e); (f) Total thermal cloud obtained after aligning 9 partial clouds, which covers the whole scene.

Table 3 shows a summary of the TPCs obtained, the number of points, the time spent per scan and the resolution used. In addition, ambient temperature (Celsius degree) and humidity data (in %) were recorded for 1 week. These data were stored in a common data environment together with point clouds and a set of IR photographs, thus complementing the monitoring of the scenario. Table 4 shows relevant information about sensors, location, time periods, data and formats used. The methodology used to obtain the TPC does not allow them to be geolocated automatically.

Table 3

Table 3. Characteristics of point clouds taken on February 27, 2023.

Table 4

Table 4. Summary of sensors, location, time periods, data and formats for monitoring the case study space.

4 Use and integration of TPCs in BIM

The main objective, in this case, is to process a thermal point cloud in order to generate an open IFC file containing information on the geometry of the curtain wall with the associated temperature (mean value) of each of its vertical and horizontal components at a certain moment.

4.1 Data collection methodology

The following steps were followed to obtain the IFC file for the curtain wall:

1. Segmenting the total thermal cloud. The total thermal cloud (TotalTCloud.txt) of the scenario was obtained following different processes (e.g., aligning, segmentation, reduction and regularization) developed in Matlab, which can be found in Adán et al. (2023c). However, in this case, since only the curtain wall part was of interest, TotalTCloud.txt was manually segmented in CloudCompare (see Figure 3). The segmentation consisted of a range filter to select only the points on the facade between 20 °C and 28 °C, eliminating outliers and noise produced by reflections, external objects, or ventilation grilles, and adjusting the false color to this range to maximize the contrast of the facade, which was at a higher temperature than the average of 19 °C shown in TotalTCloud.txt.

2. Generating the .las file. The segmented cloud was saved in CloudCompare in.las format, generating the file TCurtainCloud.las.

3. Generating the .rvt file. TCurtainCloud.las was read in Autodesk Recap and exported in.rvt format so that it can be imported into Autodesk Revit 2020. In this step, a change in the coordinate reference system occurred. Once TCurtainCloud.rvt was imported into Autodesk Revit 2020, the geometry was manually generated, structured in segments of the “panel”, “mullion” (vertical segment) and “crossbeam” (horizontal segment), which are native Revit families. This model, which we call CurtainModel.rvt, contains, for the moment, only geometric information and the native properties of the curtain wall elements (see Figure 4).

4. Obtaining Coordinates.csv. In order to save the list of 3D coordinates of the vertices of the links and crossbars in an editable file, the Dynamo plugin was used, which generated a Coordinates.csv file to be able to automatically later assign the temperature data from the TPC.

5. Aligning and assigning temperature to curtain wall elements (mullions and crossbeams). The point cloud acquired by the terrestrial laser scanner (TCurtainCloud.las) and the point cloud derived from the Revit model vertices (Coordinates.csv) were imported into MATLAB, resulting in TCurtainCloud.mat and Coordinates.mat, respectively. To align the two datasets, a coarse registration of the curtain wall edge points to a predefined world coordinate system was first performed using rigid transformation routines. This initial step ensured a rough spatial consistency between both sources. In a second stage, a refined alignment was carried out using the well-known Iterative Closest Point (ICP) algorithm, which minimized the residual distances between corresponding surface regions of the two-point sets and yielded the final accurate registration.

6. Geometric deviations between the laser-scanner point cloud and the Revit-derived vertices were quantified from the residual point-to-surface distances after ICP convergence. The ICP iterations were stopped when either (i) the relative decrease of the root-mean-square error (RMSE) between two consecutive iterations fell below $1\times {10}^{-4}$, or (ii) a maximum of 100 iterations was reached. In our case, convergence was achieved after 27 iterations, yielding a final RMSE of 0.035 m, with a mean deviation of 0.030 m (±0.012 m) and a maximum deviation of 0.070 m over all vertices.

7. Once the two datasets were aligned, the temperature values were assigned to each vertex in Coordinates.mat using a nearest-neighbor algorithm. Subsequently, the representative average temperature of each segment was computed by averaging the temperatures of its corresponding vertices (Figure 5a). We aggregated vertex temperatures using per-element averages because IFC schemas do not support point-level scalar attributes and because excessive data density would hinder interoperability. In the case study there is a large number of elements (Figure 5b) that require the assignment of a representative temperature, and this value can only be robustly obtained by averaging multiple temperature values. The mean values and standard deviation of temperature for each of the elements of the curtain wall (Mullion, Panel and Crossbeam) vary from 22.0 °C to 24.5 °C and 0.21 to 0.41 Std, while temperature absolute values vary from 21.5 to 25.5 in the TPC (Figure 5c). This summarization is consistent with thermal-simulation workflows that use element-level properties, as surface temperatures for CFD (Computational Fluid Dynamics) simulations.

8. Generating the .ifc file. Having obtained the representative average temperature for each segment of the curtain wall, the corresponding IFC ASCII code was generated using a MATLAB script, producing the file named TCurtainModel.ifc. The code was generated following the standard IFC text-based specification, using the basic parameters: ifcName: seg**_{panel, mullion, crossbeam}, ifcExportAs: IfcWallNotDefined, ifcSpatialContainer: Ground Floor, Value (Temperature): **, and ifcPropertySetList: “Temperature”.

9. Checking the .ifc file. Finally, in order to check the quality of the resulting file in a BIM environment, TCurtainModel.ifc was imported into both BIM-Vision and Revit viewers. See Figure 6. As an example, the temperature assigned to a specific element of the curtain wall can be seen as a parameter or property.

Figure 3

Figure 3. (a) Total thermal cloud (b) Segmented cloud in CloudCompare.

Figure 4

Figure 4. (a) TPC and detail of “panel” segment tracing; (b) “mullion” tracing (vertical segment); (c) “crossbeam” element tracing (horizontal segment); (d) Final curtain wall model made in Revit.

Figure 5

Figure 5. (a) Image of the original point cloud and the misaligned Revit model due to the change in reference system produced in the import step in Recap; (b) Original point cloud and the Revit model after alignment; (c) Visualization of the thermal cloud and the overexposed aligned Revit model.

Figure 6

Figure 6. (a) TCurtainModel.ifc imported into Revit. The temperature assigned to the blue element is seen in the Parameters section. (b) TCurtainModelT.ifc imported into BIMVision. The temperature assigned to the earlier element is now seen in the Property section.

4.2 Discussion about the software tools used

The software Matlab, CloudCompare, Recap, Revit, BIMVision and Dynamo were used in this process. A brief discussion about the software used can help the reader to better understand the methodology defined in the previous section.

• Use of Matlab. Matlab was used in several phases of the process, essentially as a programming language. First, the point cloud was generated in a native.txt format so that it was accessible to other commercial software. Later, the temperature assignment to each vertex of the segments created in the curtain wall model was programmed. Finally, the.ifc format was generated in Matlab, containing the coordinates of the vertices and the assigned temperature of each segment of the curtain wall.

• Use of CloudCompare. CloudCompare was used to perform a manual segmentation of the total cloud to keep the required part of the curtain wall and to export it in.las format.

• Use of Recap and Revit: Recap was used as intermediate software to move from.las to.rvt formats, while Revit was used as a modelling tool for the curtain wall. As mentioned above, the point cloud served as a guide to manually trace the curtain wall, structured in segments (parallelepipeds) of three types: panel, mullion and crossbeams. At the end of the process, the final.ifc file containing the assigned temperature was imported into Revit to verify the validity of the file. BIMVision viewer was also used to check that the temperature was correctly assigned as an attribute of each segment of the curtain wall and visualized in an open format.

• Use of Dynamo: The Dynamo plugin was used exclusively to generate the ordered lists of the coordinates of the vertices of each segment of the curtain wall.

5 Results. Integration of TPCs and local sensor data into GIS

The objective of this section is to define a methodology to integrate TPCs and other local sensor data in a GIS environment. This methodology could be extended to other buildings on the campus during long-term monitoring periods.

The starting point of our methodology is the model in.rvt format of the building studied. This model was manually obtained in Revit from the TPC, as mentioned in Section 4.1. Step 2. This is used as a layer in the ArcGIS Pro 3.5 desktop software, in which to add other geometric elements that represent the corresponding physical sensors used. At a later stage, this BIM model is uploaded to the ArcGIS Online platform along with the layers with elements created in the desktop tool. Finally, this GIS model is uploaded to ACCA-Software’s GeoTwin platform, where the TPC is finally added as another data that can be consulted. ArcGIS Online platform serves as a centralized container where the rest of the data obtained are uploaded. All this will be discussed in this section.

5.1 Data integration methodology

Because direct import of non-geolocated LAS-based TPCs was not supported in ArcGIS Online, the GIS workflow adopts a workaround in which manual multipatch geometries receive thermal attributes derived from the TPC. The methodology carried out to integrate TPCs into GIS is summarized in the following steps:

1. Geo-referenced model of the NYUAD campus. The detailed and geo-referenced model of the NYUAD campus volumes was obtained from OpenStreetMap data and integrated into a base map using satellite images. The metadata associated with the campus buildings was filtered, retaining only the parameters that identify the different buildings and specify their function (educational, office, residential, and services, among others).

2. 3D geometric model of the studied building. The 3D model made in Autodesk Revit of the double-height space studied was designed with the help of the original TPC, obtaining the Building.rvt file (see Section 4.1. Step 3).

3. Basic GIS model. To import the Building.rvt model into ArcGISPro from Revit, an internal tool of the software was used that facilitates the conversion of data “From BIM to geodatabase.” This process allows the model to be segmented into elements categorized according to the default Revit labels (Architectural, Floors, Masses, Structural, among others). It is important to note that these categories cannot be edited manually, nor can new ones be added.

4. Inserting local sensors into the GIS model. The local sensors used were added to the space in the curtain wall and the upper floor as “3DPoint” elements in the model, each assigned an attribute “name” and “link” that corresponds to a.csv file in a cloud storage (Google Drive), where the data recorded from the ambient temperature (in Celsius degree) and contact sensors during the test period are stored.

5. Inserting temperature into the GIS. Regarding the inclusion of temperature in the curtain wall façade, this was firstly converted from a commercial BIM Revit format to a three-dimensional data structure (named “multipatch”) to which custom attributes can be added and then visualized using a colour scale (Figure 7). An attribute called “TEMP” was created to store the scalar values previously obtained from the TPM through a manual thermal segmentation process by values performed in CloudCompare (Figure 8). Two different “multipatches” were created to examine how to insert data from two different readings taken at different times. Figure 9 also provides a legend showing how these temperature values are organized in the BIM-GIS model.

Figure 7

Figure 7. Multipatch object in ArcGIS Pro Scene Viewer, where a custom scale is added to visualize the temperature.

Figure 8

Figure 8. Interface of CloudCompare software showing manual filtering by temperature values of the TPC.

Figure 9

Figure 9. View in ArcGIS Online of the GIS model with the legend corresponding to the temperature values of the curtain wall and other elements on the satellite image.

The proposed TPC to GIS thermal model is available for consultation at Ramón-Constantí (2024) on the Esri online platform.

1. Inserting reference to TPC into the GIS models. Complementarily, and in order to visualize the TPC in GIS, a “3D geometric element” type icon was created with a link to the Sketchfab platform, where one of the building’s TPCs was uploaded. Since it was not possible to integrate the TPC directly using a.las file in ArcGIS Pro or its web version, this TPC can be viewed at the following link: https://skfb.ly/oPDX9.

2. Inserting TPC into ACCASoftware’s GeoTwin tool. The aforementioned GIS model has been used as a model and base map in ACCASoftware’s GeoTwin tool, where the thermal point cloud obtained from the case study has been added in.las format after being filtered for outliers. However, there are still problems in the visualization of this type of thermal models, such as the impossibility of placing scales that allow selecting a point and obtaining the value of its attribute (in our case, the temperature) natively in the application itself, without having to manually create a legend. In this case, the platform allows loading a.las model of the point cloud and geolocating it on the map (Figure 10) but does not allow interaction with it beyond being able to view it (rotate the view, zoom in and out).

Figure 10

Figure 10. View of the model in ACCASoftware’s GeoTwin tool.

5.2 Discussion about the software tools used

The tools used in this approach include Autodesk Revit, Esri ArcGIS Pro 3.5, ArcGIS Pro Online, and GeoTwin from ACCASoftware. As in the previous section, a brief discussion about the software used is included in this section. Several important details and valuable extra information can help the reader to better understand the methodology defined in this process.

The tools were used thanks to the technical support provided to the university, along with their advanced features compared to the open-source alternatives available. The tools used include not only data visualization but also the availability of toolboxes with various functionalities for the implementation of digital twins and the integration of data from the different external sources that were collecting data while the thermal point clouds were being taken.

• Use of Autodesk Revit: Autodesk Revit was used to manually obtain the geometric model from the original TPC and to add location data of the building (Figure 11a). The software was used to achieve a Level of Detail (LOD) 4 of the building studied, according to the City Geography Markup Language (CityGML) standard, which includes the interior distribution of the spaces. There are differences between the GIS and BIM LOD, as shown in Figure 11b. A LOD 4 in BIM (Figure 11c) might include very specific details, such as individual wall components or fixtures, whereas a LOD 4 in CityGML includes the interior layout of the building, but often with less construction detail than a BIM model at the same level. This can lead to inconsistencies when trying to integrate or compare data between the two systems.

• Use of Esri ArcGIS Pro: Point clouds can be included in the desktop tools as.las files, but, so far, their display is limited, and there are important restrictions when working with this type of file in this environment (for example, colour scale values cannot be chosen, clouds cannot be clipped, different clouds cannot be superimposed), as shown in Figure 12. The resulting.rvt model was exported directly to ArcGIS Pro on the desktop in order to serve as a base model where the local sensors can be placed as geometric elements in ArcGIS as was explained in Section 4.1 Methodology.

• Use of Esri ArcGIS Pro Online: To contextualize the model within the campus, open data from OpenStreetMap (OSM) was incorporated into the workflow. This complementary information was essential to accurately model the campus layout and locate the monitored buildings within it. The integration of these tools allows for a comprehensive and accurate representation of the built environment, as seen in Figure 13a, facilitating detailed analysis within the framework of thermal digital twins for better decision-making.

• ArcGIS Pro Online software was used to integrate all the generated information and upload it to the Esri online platform, allowing data to be centralized and remote access to be facilitated. On this platform, the Building Explorer interactive viewer, included in the Indoors plugin, was used, offering tools to explore and manipulate the created BIM-GIS models. Users can change viewpoints, adjust the scale (from a global view to specific building details), activate or deactivate floors, and visualize data from the sensors. This allows for interaction with the models, as shown in Figure 13b.

• Use of ACCASoftware’s GeoTwin: ACCASoftware’s GeoTwin tool was used as an alternative to display the TPC on an online platform different from Esri’s ArcGIS, where data in different formats can be included. In this case, the Geotwin application was used, which is designed to load other types of data onto a georeferenced map based on ArcGIS that this platform does not fully display, such as point clouds or IFC models that can be updated continuously. The TPC was loaded onto this platform using the model created in ArcGIS, as shown in Figure 14.

Figure 11

Figure 11. (a) 3D model made in Autodesk Revit of the double-height space studied and the façade of the interior courtyard.; (b) CityGML LOD, where LOD0 refers to plan views and LOD4 covers interior layout and furniture; (c) BIM LOD, where LOD200 is simple volumes and LOD400 includes the installations.

Figure 12

Figure 12. View of a.las file with “Intensity” values corresponding to the scalar value of the temperature at each point in ArcGIS Pro desktop.

Figure 13

Figure 13. (a) Model view alongside data provided by OpenStreetMap in ArcGIS Online; (b) Model view with Esri online viewer.

Figure 14

Figure 14. ACCASoftware Geotwin platform interface with TPC on the geolocated model.

6 Limitations

Although based on a single case study, the workflow exposes recurring interoperability gaps that are likely to arise in similar architectural contexts. The implementation of methodologies for integrating thermal models in the construction sector faces several significant limitations that affect its efficiency and usability within professional settings in the AEC industry. In summary, these are:

1. Manual modelling dependency. The TPC processing requires specific programming skills, but the rest of the process remains manual and repetitive, involving the need to manually model over the point cloud to obtain the vertices of key construction elements and then use these coordinates within the thermal point cloud to define volumes. The average temperatures of the elements must be calculated through an automated programming process. Particularly in the case of buildings with irregular or non-Manhattan geometries, a great deal of manual work is required.

2. Absence of automatic geolocation in the TPC. The TPC loses its geolocation due to the way it is processed, so it is necessary to manually geolocate this thermal data in BIM and GIS software and platforms.

6.1 Working in complex buildings and structures

Working with such complex or large-scale buildings also continues to be a future challenge, as it requires lengthy data acquisition times, a significant deployment of computing resources, and a hard processing of the data. Additionally, numerous manual data segmentation and cleaning processes (outliers) are necessary. Although some scripting is needed for the IFC generation stage, the majority of operations remain visual and do not require advanced programming. The scripting components represent isolated bottlenecks that are not conceptually intrinsic to the methodology and are a target for future automation. The methodology is presented as a proof of concept demonstrating where automation is possible and where current tools still require manual effort.

1. Inability of GIS to handle non-standard point clouds. There is a major limitation regarding the capabilities of the platforms and software used (file size limitations, supported formats, how the models can be visualized…), both open-source and commercial, which are intended for the creation of digital models or the visualization of building data. In the case of software designed to work with BIM, it's possible to import and visualize point clouds with colour but not manipulate the data (add or change values, trim the cloud, add new clouds…). For thermal clouds, the visualization doesn’t go beyond a false colour-temperature display.

2. Commercial tool reliance. In the case of GIS platforms, they are not equipped to handle non-standard, non-geolocated LIDAR point clouds without the metadata required by software or platforms such as ArcGIS. This workaround does not constitute full TPC integration. Its purpose is instead to demonstrate practical feasibility under current platform constraints.

7 Conclusion

The proposed proof of concept has proven to be effective addressing various aspects that had not been previously addressed in similar research, such as the use of thermal point clouds in large-scale buildings with several floors, the presence of a mostly glazed façade with complex geometry, the implementation of a sensor network in the as-is building and the subsequent integration of the collected data into a comprehensive BIM-GIS model that will serve as a basis for future developments. Temperatures have also been assigned to all the elements that make up the façade, which allows the creation of a data structure for a geometric-thermal model that can be consulted in the cloud.

This article also presents two TPC applications in the early stages of development for inclusion in architectural digital twins. TPCs are useful tools for understanding the thermal behaviour of an as-is space or infrastructure, and offer a comprehensive view for agents involved in its preventive maintenance. Although, in a limited way, it is possible to work with these models in common formats in the architecture, engineering and construction industry, such as the IFC format. Although the use of TPC in GIS environments is interesting, there are still important limitations today in its implementation as a useful methodology for professionals in the AECO sector.

1. Regarding the sensing system: the employed method faces challenges when performing laser scans in short periods of time, which implies the need to generate TPC for long periods, especially in very large spaces that require between 1 and 3 h of scanning. Likewise, the lack of automation in the acquisition and processing methodology limits efficiency, especially in cases of buildings with multiple floors, since the lack of automated tools implies the need for manual processing in subsequent steps by a specialist. This stage involves manually obtaining average temperature values, which must then be applied to the BIM-GIS model.

2. Regarding the use of software and dependence on commercial software: most of the tools currently used are commercial, such as Autodesk Revit and Esri ArcGIS Pro, although free software tools are also used in intermediate processes, such as CloudCompare. There is a need to integrate more open-source tools into the entire workflow, such as Blender or QGIS, although this requires adequate programming skills that are not common for professionals engaged in preventive maintenance or architectural design. The physical deployment of media is also a challenge, as data collection in large-scale spaces demands significant resource mobilization, and the connection of sensors to power sources and wireless networks to download the data obtained can be limited, resulting in a lack of these for a complete quantitative analysis.

3. GIS limitations when handling point clouds: Regarding thermal visualization, although thermal data has been collected, its visualization and analysis in BIM and GIS environments still have limitations. The existing literature on thermal data visualization in BIM-GIS is scarce or non-existent, highlighting the need to develop and apply more advanced techniques in this field. Finally, the use of GIS has proven to be valuable in monitoring indoor spaces, especially for the precise location of elements such as sensors and the segmentation of buildings into smaller units such as structural elements or independent rooms, facilitating the creation of digital twins.

4. Lack of automatic geolocation of the TPCs: Although GIS is valuable for sensor location and building segmentation, there are still limitations in its ability to efficiently manage TPCs. Moreover, the lack of inherent geolocation data in the TPCs can limit their relevance in certain digital twin applications.

8 Future works

Progress in the integration of thermal models in the construction context requires addressing the identified limitations and improving the efficiency of several of the processes performed in the case studies. An important focus is the development of more automated methods, accessible to AECO industry stakeholders without specialized computer skills, for the conversion of TPCs into BIM models. It is important that these methods can deal with complex geometries and the diversity of elements and structures present in most current construction environments. Automating thermal Scan-to-BIM would not only streamline processes but could also make the integration of thermal models more accessible to a wider majority of professionals in the sector. In addition, the integration of thermal data from different sensor networks over time into the specific data structures of BIM formats needs to be explored. This integration will allow for continuous and detailed analysis of the thermal behaviour of buildings, which in turn will facilitate the identification of areas for energy improvement and the optimization of facility performance and comfort.

Regarding the integration of TPC into GIS, similar challenges are encountered in terms of data visualization and processing. Some of these are the precise geolocation of the positions from which the thermal point clouds have been taken, together with the inclusion of other geolocated thermal data, such as point clouds obtained from drone flights, and improvement in importing and integrating data coming from sensors of different types.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

AR-C: Data curation, Formal Analysis, Methodology, Software, Validation, Visualization, Writing – original draft. FC: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – review and editing. BGdS: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – review and editing. AA-O: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Software, Supervision, Validation, Writing – review and editing.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This publication is as part of the PID2022-141275OB-C21, CIN/AEI/10.13039/501100011033/FEDER, UE and SBPLY23180225000113 projects co-funded by FEDER (European Regional Development Fund), and 2022-GRIN-34261 project funded by Plan Propio de Investigación at Universidad Castilla-La Mancha. Part of this research was partially supported by various centers at NYU Abu Dhabi (NYUAD), specifically the Center for Sand Hazards and Opportunities for Resilience, Energy, and Sustainability (SHORES) and the Center for Interacting Urban Networks (CITIES). These centers are funded by Tamkeen under the NYUAD Research Institute Awards CG013 and CG001, respectively.

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.

Footnotes

^{Abbreviations:}AECO: Architecture, Engineering, Construction and Operation; TPM: Thermal Proprietary Models; TPC: Thermal point clouds; BIM: Building Information Models; BEM: Building Energy Model; BES: Building Energy Simulation; BMS: building management system; GIS: Geographic Information System; IR: infrared; CityGML: City Geography Markup Language; LOD: Level of Detail; OSM: OpenStreetMap; UAV: Unmanned Aerial Vehicle; CFD: Computational Fluid Dynamics; IFC: Industry Foundation Classes; TLS: terrestrial laser scanning; VPI: visual programming interface.

References

Adamopoulos, E., Patrucco, G., Volinia, M., Girotto, M., Rinaudo, F., Giulio Tonolo, F., et al. (2021). “3D thermal mapping of architectural heritage: up-to-date workflows for the production of three-dimensional thermographic models for built heritage NDT,” in Lecture notes in computer science (including subseries lecture notes in artificial intelligence and lecture notes in bioinformatics) (Cham: Springer), 26–37. doi:10.1007/978-3-030-73043-7_3

CrossRef Full Text | Google Scholar

Adán, A., Quintana, B., García Aguilar, J., Pérez, V., and Castilla, F. J. (2020). Towards the use of 3D thermal models in constructions. Sustain. Switz. 12 (20), 1–13. doi:10.3390/su12208521

CrossRef Full Text | Google Scholar

Adán, A., Ramón, A., Vivancos, J., Vilar, A., and Aparicio-Fernández, C. (2023c). Automatic generation of as-is BEM models of buildings. J. Build. Eng. 73, 106865. doi:10.1016/J.JOBE.2023.106865

CrossRef Full Text | Google Scholar

Adán, A., López-Rey, A., and Ramón, A. (2023a). “Obtaining 3D dense thermal models of interiors of buildings using Mobile robots,” in ROBOT2022: fifth iberian robotics conference (Springer International Publishing), 3–14. doi:10.1007/978-3-031-21065-5_1

CrossRef Full Text | Google Scholar

Adán, A., López-Rey, A., and Ramón, A. (2023b). Robot for thermal monitoring of buildings. Automation Constr. 154, 105009. doi:10.1016/j.autcon.2023.105009

CrossRef Full Text | Google Scholar

Alba, M. I., Barazzetti, L., Scaioni, M., Rosina, E., and Previtali, M. (2011). Mapping infrared data on terrestrial laser scanning 3D models of buildings. Remote Sens. 3 (9), 1847–1870. doi:10.3390/rs3091847

CrossRef Full Text | Google Scholar

Albourae, A. T., Armenakis, C., and Kyan, M. (2017). Architectural heritage visualization using interactive technologies. Int. Archives Photogrammetry, Remote Sens. Spatial Inf. Sci. - ISPRS Archives, 7–13. doi:10.5194/isprs-archives-XLII-2-W5-7-2017

CrossRef Full Text | Google Scholar

Amirebrahimi, S., Rajabifard, A., Mendis, P., and Ngo, T. (2016). A BIM-GIS integration method in support of the assessment and 3D visualisation of flood damage to a building. J. Spatial Sci. 61 (2), 317–350. doi:10.1080/14498596.2016.1189365

CrossRef Full Text | Google Scholar

Antón, D., and Amaro-Mellado, J. L. (2021). Engineering graphics for thermal assessment: 3D thermal data visualisation based on infrared thermography, GIS and 3D point cloud processing software. Symmetry 13 (2), 1–20. doi:10.3390/sym13020335

CrossRef Full Text | Google Scholar

Back, Y., Kumar, P., Bach, P. M., Rauch, W., and Kleidorfer, M. (2023). Integrating CFD-GIS modelling to refine urban heat and thermal comfort assessment. Sci. Total Environ. 858, 159729. doi:10.1016/j.scitotenv.2022.159729

PubMed Abstract | CrossRef Full Text | Google Scholar

Bottaccioli, L., Aliberti, A., Ugliotti, F., Patti, E., Osello, A., Macii, E., et al. (2017). “Building energy modelling and monitoring by integration of IoT devices and building information models,” in Proceedings - International computer software and applications conference (IEEE), 914–922. doi:10.1109/COMPSAC.2017.75

CrossRef Full Text | Google Scholar

Bracht, M. K., Melo, A. P., and Lamberts, R. (2021). A metamodel for building information modeling-building energy modeling integration in early design stage. Automation Constr. 121, 103422. doi:10.1016/j.autcon.2020.103422

CrossRef Full Text | Google Scholar

Deng, Y., Cheng, J. C. P., and Anumba, C. (2016). Mapping between BIM and 3D GIS in different levels of detail using schema mediation and instance comparison. Automation Constr. 67, 1–21. doi:10.1016/j.autcon.2016.03.006

CrossRef Full Text | Google Scholar

Desogus, G., Frau, C., Quaquero, E., and Rubiu, G. (2023). From building information model to digital twin: a framework for building thermal comfort monitoring, visualizing, and assessment. Buildings 13 (8), 1971. doi:10.3390/buildings13081971

CrossRef Full Text | Google Scholar

Dlesk, A., Strogonov, V., Vach, K., and Pollert, J. (2023). Integration of spatially oriented data from iot to facility management and bim. Int. Archives Photogrammetry, Remote Sens. Spatial Inf. Sci. - ISPRS Archives, 31–36. doi:10.5194/isprs-archives-XLVIII-5-W2-2023-31-2023

CrossRef Full Text | Google Scholar

Escudero, P. A., González-López, C., and García-Valldecabres, J. L. (2024). Methodology for data integration in 3D-HBIM digital models. Case Study The Holy Chalice Chapel Valencia Cathedr. doi:10.20365/DISEGNARECON.32.2024.11

CrossRef Full Text | Google Scholar

García-Valldecabres, J. L., Liu, J., Willkens, D. S., Escudero, P. A., López-González, C., Cortés Meseguer, L., et al. (2023). Development of a virtual itinerary with hbim and gis. Int. Archives Photogrammetry, Remote Sens. Spatial Inf. Sci., 645–652. doi:10.5194/isprs-archives-xlviii-m-2-2023-645-2023

CrossRef Full Text | Google Scholar

Göçer, Ö., Hua, Y., and Göçer, K. (2016). A BIM-GIS integrated pre-retrofit model for building data mapping. Build. Simul. 9 (5), 513–527. doi:10.1007/s12273-016-0293-4

CrossRef Full Text | Google Scholar

Hong-mei, Z., Cheng-hu, Z., Wei-qiang, G., and Jin-cai, D. (2001). Spatial distribution characteristics of urban thermal conditions: application of GIS and remote sensing. J. Geogr. Sci. 11 (3), 261–270. doi:10.1007/bf02892309

CrossRef Full Text | Google Scholar

Hou, Y., Volk, R., Chen, M., and Soibelman, L. (2021). Fusing tie points’ RGB and thermal information for mapping large areas based on aerial images: a study of fusion performance under different flight configurations and experimental conditions. Automation Constr. 124, 103554. doi:10.1016/j.autcon.2021.103554

CrossRef Full Text | Google Scholar

Isikdag, U., Zlatanova, S., and Underwood, J. (2013). A BIM-oriented model for supporting indoor navigation requirements. Comput. Environ. Urban Syst. 41, 112–123. doi:10.1016/j.compenvurbsys.2013.05.001

CrossRef Full Text | Google Scholar

Kamel, E., and Memari, A. M. (2019). Review of BIM’s application in energy simulation: tools, issues, and solutions. Automation Constr. 97, 164–180. doi:10.1016/j.autcon.2018.11.008

CrossRef Full Text | Google Scholar

Lagüela, S., Díaz-Vilariño, L., Roca, D., and Armesto, J. (2014). “Aerial oblique thermographic imagery for the generation of building 3D models to complement geographic information systems,” in Proceedings of the 2014 international conference on quantitative InfraRed thermography (Bordeaux, France: QIRT Conference). doi:10.21611/qirt.2014.041

CrossRef Full Text | Google Scholar

Lin, D., Jarzabek-Rychard, M., Tong, X., and Maas, H. G. (2019). Fusion of thermal imagery with point clouds for building façade thermal attribute mapping. ISPRS J. Photogrammetry Remote Sens. 151, 162–175. doi:10.1016/j.isprsjprs.2019.03.010

CrossRef Full Text | Google Scholar

López-González, C., and García-Valldecabres, J. (2023). The integration of HBIM-SIG in the development of a virtual itinerary in a historical centre. Sustainability 15 (18), 13931. doi:10.3390/su151813931

CrossRef Full Text | Google Scholar

Mutani, G., Delmastro, C., Gargiulo, M., and Corgnati, S. P. (2016). Characterization of building thermal energy consumption at the urban scale. Energy Procedia 101 (September), 384–391. doi:10.1016/j.egypro.2016.11.049

CrossRef Full Text | Google Scholar

Niu, S., Pan, W., and Zhao, Y. (2015). A BIM-GIS integrated web-based visualization system for low energy building design. Procedia Eng. 121, 2184–2192. doi:10.1016/j.proeng.2015.09.091

CrossRef Full Text | Google Scholar

Oti, A. H., Kurul, E., Cheung, F., and Tah, J. (2016). A framework for the utilization of building management system data in building information models for building design and operation. Automation Constr. 72, 195–210. doi:10.1016/j.autcon.2016.08.043

CrossRef Full Text | Google Scholar

Previtali, M., Erba, S., Rosina, E., Redaelli, V., Scaioni, M., and Barazzetti, L. (2012). “Generation of a GIS-based environment for infrared thermography analysis of buildings,” in Infrared remote sensing and instrumentation. Editors M. Strojnik, and G. Paez, XX, 85110U. doi:10.1117/12.930050

CrossRef Full Text | Google Scholar

Rahhal, A., Matthys, C., Ben Rajeb, S., and Leclercq, P. (2019). “Limitations of using digital BIM models to carry out thermal analysis,” in Fassi 2019: fifth international conference on fundamentals and advances in software systems integration. [Preprint].

Google Scholar

Ramón-Constantí, A., Adán Oliver, A., Quintana Galera, B., Castilla Pascual, F. J., and Pérez-Andréu, V. J. (2022a). “Technology and methodology for obtaining omnidirectional 3D thermal point clouds models of buildings,” in 26th international congress on project management and engineering terrassa.

Google Scholar

Ramón-Constantí, A., Adán-Oliver, A., and Castilla-Pascual, F.-J. (2022b). Thermal point clouds of buildings: a review. Energy Build. 274, 112425. doi:10.1016/j.enbuild.2022.112425

CrossRef Full Text | Google Scholar

Ramón-Constantí, A., Adán-Oliver, A., Castilla-Pascual, F. J., and Pérez-Andreu, V. (2024). An experimental methodology for the calibration of indoor building environment models using thermal point clouds and CFD simulation. Adv. Build. Energy Res. 18 (3), 261–294. doi:10.1080/17512549.2024.2358923

CrossRef Full Text | Google Scholar

Tomchinskaya, T., and Galanina, M. (2020). “University management system based on bim-gis-technologies,” in CEUR workshop proceedings. doi:10.51130/graphicon-2020-2-4-26

CrossRef Full Text | Google Scholar