Introduction
Digital Twin technology is becoming an important part of digital transformation in the Architecture, Engineering, and Construction industry. A Digital Twin in architecture integrates Building Information Modeling with real-time sensor data to simulate, monitor, and analyze a construction asset. While current Vietnamese research often prioritizes large projects such as commercial centers and hospitals, applications for houses and compact offices remain underexplored. This study addresses that gap with a conceptual framework for an IoT-integrated Digital Twin tailored to small-scale smart buildings in Vietnam.
Background and literature review
The proposed DTiA approach connects sensor data collection, cloud-based transmission, and real-time BIM integration. Prior work has examined BIM for energy performance analysis, sensor-linked building models, and BIM-based Digital Twin platforms such as SPHERE. These studies establish the value of integrating physical sensing with semantic and spatial building information, while also highlighting the need for accessible solutions suited to smaller projects.
Levels of Digital Twin development
The framework distinguishes three levels of digital replica. A Digital Model is a static representation updated through manual inputs. A Digital Shadow adds automated one-way data flow from the physical building to the digital model for real-time monitoring. A Digital Twin adds bidirectional communication, allowing the digital model to send control commands back to the physical system. The proof-of-concept in this study targets the Digital Shadow level.
Proposed conceptual framework
The framework treats a Digital Twin as a continuous cyber-physical ecosystem rather than a static 3D representation. It uses a dual-axis structure: a vertical five-layer technology architecture from physical sensors to application interfaces, and a horizontal lifecycle strategy spanning design, construction, and operation. Together, these axes provide a scalable roadmap for deploying IoT-integrated BIM systems in small-scale built environments.
Five-layer technology architecture
Layer 1, Physical Assets and Sensors, includes the building, its spaces and components, and an IoT network using cost-effective devices such as ESP32 microcontrollers with temperature, humidity, and particulate matter sensors. Layer 2, Data and Communication, transmits measurements through lightweight publish/subscribe protocols, particularly MQTT. Layer 3, Digital System, uses BIM platforms such as Autodesk Revit as dynamic visual databases, with Dynamo or custom APIs mapping sensor streams to architectural objects and spaces. Layer 4, Intelligence, performs normalization, noise filtering, and historical storage, with future extensions for anomaly detection and predictive maintenance. Layer 5, Application and Interface, provides BIM visualization and web dashboards for architects, engineers, and facility managers.
Lifecycle integration
During the design phase, historical or analogous microclimatic data can support spatial organization, ventilation strategies, material selection, and indoor environmental quality. During construction, site sensors can monitor dust, temperature, safety conditions, and progress against planned schedules. During operation and maintenance, continuous environmental and energy data can support facility management, early fault detection, scenario planning, and predictive maintenance when the Intelligence layer is fully developed.
Illustrative proof-of-concept
The experimental system uses an ESP32 with DHT22 temperature and humidity sensing and fine particulate matter sensors in a simulated building space. MQTT carries measurements to a Node.js server for processing and storage. Autodesk Revit provides the digital representation, while Dynamo scripts periodically retrieve processed data and update parameters of corresponding 3D rooms. A Next.js dashboard visualizes current and historical metrics. A 72-hour test reports an average end-to-end latency of 2.4 seconds, a peak of 3.1 seconds, and a packet delivery ratio of 99.1 percent. These results support the technical feasibility of the proposed Digital Shadow implementation.
Cost-effectiveness
The study estimates custom ESP32-based sensor nodes at approximately 12 to 15 US dollars per unit, representing a substantial reduction compared with industrial BACnet or KNX sensors. Although the current Intelligence layer focuses on normalization rather than advanced machine learning, the proof-of-concept indicates that the five-layer architecture can provide an economically accessible foundation for small-scale smart buildings in Vietnam.
Conclusion
This study formulates a conceptual framework for integrating Digital Twin technology into small-scale smart buildings. Its main contribution is the combination of a vertical five-layer architecture with a horizontal lifecycle integration strategy. The ESP32, MQTT, and Autodesk Revit proof-of-concept reaches the Digital Shadow level through reliable one-way synchronization of environmental conditions. Future research will extend the Intelligence layer with predictive machine learning and bidirectional control loops, moving toward a more autonomous Digital Twin ecosystem.
Acknowledgements
This research is funded by Ho Chi Minh City University of Technology (HCMUT), VNU-HCM under grant number SVHT-2025-KTXD-02. The authors acknowledge Ho Chi Minh City University of Technology (HCMUT), VNU-HCM for supporting this study.
References
Tao, F., Zhang, H., Liu, A., and Nee, A. Y. C. (2018). Digital Twin in Industry: State-of-the-Art. IEEE Transactions on Industrial Informatics, 15(4), 2405-2415. Kritzinger, W., Karner, M., Traar, G., Henjes, J., and Sihn, W. (2018). Digital Twin in manufacturing: A categorical literature review and classification. IFAC-PapersOnLine, 51(11), 1016-1022. Khajavi, S. H., Motlagh, N. H., Jaribion, A., Werner, L. C., and Holmstrom, J. (2019). Digital Twin: Vision, benefits, boundaries, and creation for buildings. IEEE Access, 7, 147406-147419. Boje, C., Guerriero, A., Kubicki, S., and Rezgui, Y. (2020). Towards a semantic construction Digital Twin: Directions for future research. Automation in Construction, 114, 103179. Jones, D., Snider, C., Nassehi, A., Yon, J., and Hicks, B. (2020). Characterising the Digital Twin: A systematic literature review. CIRP Journal of Manufacturing Science and Technology, 29, 36-52. Abanda, F. H., and Byers, L. (2016). An investigation of the impact of building orientation on energy consumption in a domestic building using emerging BIM. Energy, 97, 517-527. Alonso, R., Borras, M., Koppelaar, R. H. E. M., Lodigiani, A., Loscos, E., and Yontem, E. (2019). SPHERE: BIM Digital Twin Platform. Proceedings, 20(1), 9. Afsari, K., Florez, L., Maneke, E., and Afkhamiaghda, M. (2019). An Experimental Investigation of the Integration of Smart Building Components with Building Information Model (BIM). Proceedings of the 36th ISARC.