Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning
As the world moves toward a low-carbon future, a key challenge is improving buildings’ energy performance while maintaining occupant thermal comfort. Emerging digital tools such as the Internet of Things (IoT) and Building Information Modeling (BIM) offer significant potential, enabling precise moni...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Buildings |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-5309/15/10/1584 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850127205945835520 |
|---|---|
| author | Fahad Iqbal Shayan Mirzabeigi |
| author_facet | Fahad Iqbal Shayan Mirzabeigi |
| author_sort | Fahad Iqbal |
| collection | DOAJ |
| description | As the world moves toward a low-carbon future, a key challenge is improving buildings’ energy performance while maintaining occupant thermal comfort. Emerging digital tools such as the Internet of Things (IoT) and Building Information Modeling (BIM) offer significant potential, enabling precise monitoring and control of building systems. However, integrating these technologies into a unified Digital Twin (DT) framework remains underexplored, particularly in relation to thermal comfort. Additionally, real-world case studies are limited. This paper presents a DT-based system that combines BIM and IoT sensors to monitor and control indoor comfort in real time through an easy-to-use web platform. By using BIM spatial and geometric data along with real-time data from sensors, the system visualizes thermal comfort using a simplified Predicted Mean Vote (sPMV) index. Furthermore, it also uses a hybrid machine learning model that combines Facebook Prophet and Long Short-Term Memory (LSTM) to predict the future indoor environmental parameters. The framework enables Model Predictive Control (MPC) while providing building managers with a scalable tool to collect, analyze, visualize, and optimize thermal comfort data in real time. |
| format | Article |
| id | doaj-art-a273e4087c724889b5b55f934b13df34 |
| institution | OA Journals |
| issn | 2075-5309 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Buildings |
| spelling | doaj-art-a273e4087c724889b5b55f934b13df342025-08-20T02:33:43ZengMDPI AGBuildings2075-53092025-05-011510158410.3390/buildings15101584Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine LearningFahad Iqbal0Shayan Mirzabeigi1Department of Sustainable Resources Management, State University of New York College of Environmental Science and Forestry, Syracuse, NY 13210, USADepartment of Sustainable Resources Management, State University of New York College of Environmental Science and Forestry, Syracuse, NY 13210, USAAs the world moves toward a low-carbon future, a key challenge is improving buildings’ energy performance while maintaining occupant thermal comfort. Emerging digital tools such as the Internet of Things (IoT) and Building Information Modeling (BIM) offer significant potential, enabling precise monitoring and control of building systems. However, integrating these technologies into a unified Digital Twin (DT) framework remains underexplored, particularly in relation to thermal comfort. Additionally, real-world case studies are limited. This paper presents a DT-based system that combines BIM and IoT sensors to monitor and control indoor comfort in real time through an easy-to-use web platform. By using BIM spatial and geometric data along with real-time data from sensors, the system visualizes thermal comfort using a simplified Predicted Mean Vote (sPMV) index. Furthermore, it also uses a hybrid machine learning model that combines Facebook Prophet and Long Short-Term Memory (LSTM) to predict the future indoor environmental parameters. The framework enables Model Predictive Control (MPC) while providing building managers with a scalable tool to collect, analyze, visualize, and optimize thermal comfort data in real time.https://www.mdpi.com/2075-5309/15/10/1584thermal comfortmachine learningBIMIoTdigital twinfacility management |
| spellingShingle | Fahad Iqbal Shayan Mirzabeigi Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning Buildings thermal comfort machine learning BIM IoT digital twin facility management |
| title | Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning |
| title_full | Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning |
| title_fullStr | Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning |
| title_full_unstemmed | Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning |
| title_short | Digital Twin-Enabled Building Information Modeling–Internet of Things (BIM-IoT) Framework for Optimizing Indoor Thermal Comfort Using Machine Learning |
| title_sort | digital twin enabled building information modeling internet of things bim iot framework for optimizing indoor thermal comfort using machine learning |
| topic | thermal comfort machine learning BIM IoT digital twin facility management |
| url | https://www.mdpi.com/2075-5309/15/10/1584 |
| work_keys_str_mv | AT fahadiqbal digitaltwinenabledbuildinginformationmodelinginternetofthingsbimiotframeworkforoptimizingindoorthermalcomfortusingmachinelearning AT shayanmirzabeigi digitaltwinenabledbuildinginformationmodelinginternetofthingsbimiotframeworkforoptimizingindoorthermalcomfortusingmachinelearning |