=Paper=
{{Paper
|id=Vol-3080/paper11
|storemode=property
|title=An Integrated BIM-IoT approach to support energy monitoring
|pdfUrl=https://ceur-ws.org/Vol-3080/11.pdf
|volume=Vol-3080
|authors=Caterina Gabriella Guida,Brij B. Gupta,Angelo Lorusso,Francesco Marongiu,Domenico Santaniello,Alfredo Troiano
|dblpUrl=https://dblp.org/rec/conf/syscom/GuidaGLMST21
}}
==An Integrated BIM-IoT approach to support energy monitoring==
https://ceur-ws.org/Vol-3080/11.pdf
An Integrated BIM-IoT approach to support energy monitoring
Caterina Gabriella Guida1, Brij B. Gupta2, Angelo Lorusso3, Francesco Marongiu3,
Domenico Santaniello3 and Alfredo Troiano4
1
DICiv, University of Salerno, Salerno, Italy
2
National Institute of Technology Kurukshetra, Kurukshetra, Haryana 136119, India & Asia University,
Taichung 413, Taiwan & Staffordshire University, Stoke-on-Trent ST4 2DE, UK
3
DIIn, University of Salerno, Salerno, Italy
4
NetCom Engineering, Naples, Italy
Abstract
BIM, Building Information Model, is considered the 3D representation of an artifact and its
characteristics such as geometry, spatial relationships, and geographical information to
support integrated design. With recent developments in BIM, there has been a shift from
simple 3D to interact with virtual and augmented reality. This innovation drives
improvements in work productivity, home comfort, and entertainment, common goals of the
Internet of Things (IoT). Therefore, 3D and sensors can integrate through data captured in a
BIM model, an environment that lends itself well to the visualization of the results of
Machine Learning operations. This paper proposes a methodology that allows data
visualization and representation from sensors within a BIM model to support design
decisions that fall under different disciplines. The research focuses on a real case study of a
university classroom that includes several sensors capable of recording data that feed a
database based on the predictive/decisional phase developed through Machine Learning
techniques to optimize electrical consumption. The proposed methodology integrates an IoT
cloud platform that allows the optimal management and monitoring of electricity
consumption in a public environment through a model updated in real-time.
Keywords 1
Building Information Model, Internet of Things, Energy Monitoring, Data Analysis
1. Introduction
BIM (Building Information Modeling) is widely used in the construction industry, focusing mainly
on building systems and components. Building applications through BIM create advantages in various
applications, for example, forecasting and organizing maintenance work, energy management,
sustainability assessment, and life cycle costs [1], [2], [16]. With the advancement of technological
innovation, the amount of data that can be stored has grown, and this increase is partly due to the use
of technology at a more affordable cost allowing the development of devices connected to the Internet
continuously: this technology is called the Internet of Things, which refers to devices able to
exchange information autonomously [3], [4]. Thanks to the development of this technology, the IoT
has emerged from its embryonic stage as a revolutionary technology supporting a fully integrated
internet designed to handle different application scenarios such as smart industries, smart cities, smart
buildings, and smart healthcare [18-20]. Indeed, a crucial role of the IoT is to create a digital copy of
reality, generating digital scenarios for the management of different sectors [5], [6]. During COVID-
19 scenario the use of IoT devies are increased [24-26]. This study gives special attention to data
visualization, which allows preliminary analysis, continuous monitoring, and final processing of the
collected data using computer graphics techniques and interactive technologies. Furthermore, data
International Conference on Smart Systems and Advanced Computing (Syscom-2021), December 25–26, 2021
EMAIL: cguida@unisa.it (A. 1); gupta.brij@gmail.com (A. 2); alorusso@unisa.it (A. 3); fmarongiu@unisa.it (A. 4); dsantaniello@unisa.it
(A. 5); a.troiano@netcomgrouop.eu (A. 6)
ORCID: 0000-0003-4929-4698 (A. 2); 0000-0002-0831-5694 (A. 3); 0000-0001-5563-0411 (A. 4); 0000-0002-5783-1847 (A. 5)
©️ 2021 Copyright for this paper by its authors.
Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR Workshop Proceedings (CEUR-WS.org)
collection methods have evolved from the traditional wired transmission to open wireless
technologies such as RFID tags, embedded sensors, actuator nodes, and MEMS Micro-Electro-
Mechanical Systems [7], [8]. [17] [21-23].
2. Related Works
BIM modeling technology includes information and visualization of the building but does not have
the means to incorporate information about the surrounding environment into the model
automatically. In fact, such modeling must be supported by a plug-in for integration between reality
and the BIM model, which is the first step towards an intelligent building [9]. The digital twin can
integrate the Internet of Things, artificial intelligence, machine learning, and data analysis with spatial
network graphs. It can also use machine learning and artificial intelligence systems to process data
and produce new knowledge or predict actionable scenarios by developing collected data, creating
digital simulation models that update and change when their physical counterparts change. Therefore,
a digital twin can continuously encapsulate and update its characteristics from multiple sources [10].
Several studies have attempted to integrate sensor information and BIM information, such as
visualizations, and to build management systems, to find that data obtained from building structural
health diagnosis could not be effectively and systematically integrated with other relevant monitoring
data to support management data and proposed a dynamic parametric BIM approach using time series
sensor data to support dynamic and data-driven visualization [11]. This study integrates the
management of data collected from IoT-based sensors to a BIM model and the consequential 3D
visualization, where all sensors are represented in the BIM model and exchange information by
interacting with each other. This work describes a methodological proposal that allows the
visualization and representation in a smooth way of the data acquired by the sensors within the BIM
environment to support design decisions involving different skills from multiple disciplines [12]. The
study is based on a real case study of a university classroom comprising multiple sensors capable of
producing data that feed a database at the basis of the predictive/decision-making phase developed
through Machine Learning techniques [13]. The methodological proposal includes an IoT-based cloud
platform that assists the communication between sensors and Dynamo software to access the data,
automatically updating the information in the BIM model [14].
3. Proposed Approach
This study enhances the ability to intelligently monitor and manage a BIM model according to precise
asset steps that include a modeling technique that focuses on producing a 3D model of the case study
environments and the management of the spatial information of the same. The collection of data on an
IoT-based cloud platform allows easy and intuitive data management; the development of the
automated exchange of sensor data between sensor data collection software, parametric design
software, and the BIM modeling platform and, finally, the integration of Machine Learning
techniques for the management of the architectural space from the point of view of living comfort.
This approach can be considered in the more advanced stages of designing or managing an
environment to find the best design solution among several alternative scenarios under different
conditions. First, the basic requirements and the goal to be achieved are established. Subsequently, the
designer parameterizes the coefficients to improve the characteristics of the model. The application of
parametric design has been successfully adopted in several BIM applications as a design change
management engine; however, parametric systems have evolved into practical design tools but are not
yet considered complete BIM design applications. Dynamo is an open-source visual programming
application from Autodesk that aims to be accessible to programmers and computer language novices.
It gives users the ability to visually describe behavior, define custom logical elements, and execute
lines of code using various textual programming languages such as Python. Visual programming is a
form of coding that, unlike textual programming, does not require the compilation of code or
familiarity with a textual programming language; instead, it uses a visual interface in which the
designer connects nodes of predefined functionality. Together, these nodes form a more extensive
network of functionality capable of achieving complex goals. This approach is easier to use and
explore than programming based solely on lines of text. It makes tasks previously reserved for
experienced programmers accessible to building designers. Designers can also use the power of the
Revit API to use individual objects or families of objects in the program to perform parametric
operations. Dynamo allows users to set up automation or calculation platforms through a visual node-
based compilation interface. Designers can perform data processing and correlate structural and
geometric parameter checks.
4. Case of Study
In this study, we propose to analyze a workflow involving the collection of data through light and
lighting sensors, a detection system with cameras and energy consumption, and their integration
through a parametric control mechanism and visualization modules to facilitate management and
monitoring, with the aim of activating autonomous choice mechanisms to improve electricity
consumption due to the lighting of an environment. This workflow is developed in two architectural
phases: Visual Programming and Parametric Design. Visual programming allows the connection with
the data collection platform, the set of values and the data display mode, and the automatic integration
with the BIM work environment. On the other hand, to achieve the goal of displaying the data
obtained for improving consumption, it exploits the potential of parametric design: scale-up of the
real building with sensors through the development of the Digital Twin (BIM model) by introducing a
3D virtual space and positioning the sensors detected by an IoT-based platform, parameterizing,
according to the decision objectives related to the different scenarios, the operating rules of the
custom nodes that are implemented within the Visual Programming environment, by automatically
updating the Digital Twin. The case study focuses on reproducing a digital twin of a university
classroom. Firstly, the classroom understudy was furnished with sensors that monitor the different
parameters of its current state interconnected to a microcontroller, Arduino. The sensors used in this
study were mainly three: 1) a Light Dependent Resistor sensor placed near the external openings to
collect data about the amount of light entering the classroom; 2) video cameras, to monitor the
number of people inside a classroom and the spatial distribution within it; 3) sensor to monitor the
energy consumption of the lighting system. The acquired data is then stored on a cloud platform,
ThingsBoard [15], which allows the data to be read and visualized immediately and easily,
customizing the type of data according to the type of sensor selected or the design requirements. The
next step is the real-time visualization of the data from the sensors within the BIM model, and this
becomes possible through the use of Dynamo. Once all the data has been collected using
reinforcement machine learning, it is possible to automatically manage the intensity of the light inside
the classroom, based on the amount of light coming in from outside and weighted by the amount and
distribution of students inside the classroom to limit energy consumption and, therefore, protect the
environment. Reinforcement learning is a machine learning technique that aims to create autonomous
agents capable of choosing actions to be performed to achieve specific objectives through interaction
with the environment in which they are placed. In fact, the processor uses trial and error to solve the
problem. In this approach, the artificial intelligence receives rewards or penalties for the actions it
performs from the programmer to maximize the rewards to be received. It is up to the model to figure
out how to perform the task to maximize the reward, starting with totally random trials and ending
with sophisticated tactics. Leveraging the power of search and many trials, reinforcement learning is
the most effective way to suggest machine creativity. Finally, thanks to the dedicated API, the
ThingsBoard platform is automatically queried to capture the actual data and then, via a display,
directly to the Digital Twin in a BIM environment. The ability to use the data in real-time allows a
better study of the electrical consumption and the light intensity needed by the students inside the
classroom, becoming a tool to support design decisions to make appropriate changes in real-time,
according to the needs, and then see the effects of these choices in a practical and fast way.
Figure 1 Proposed Architecture
5. Conclusions
This work aimed to introduce a methodology that allows data management from sensors within the
BIM environment to support decisions, sometimes complex, that require interdisciplinary skills. The
case study was the monitoring and controlling lighting in a public environment such as a university
classroom that includes several sensors to reduce energy consumption due to lighting. The proposed
methodology integrates the use of an IoT-based platform, Thingsboard, which allows communication
between sensors and Dynamo software to access sensor data, automatically updating the information
contained in the BIM model. The experimental results, although preliminary, yielded promising
results. They have shown that the system can learn and manage specific actions autonomously,
supporting users. Future developments include expanding the database with other useful sensors to
refine the data, the introduction of contextual parameters that could improve the system's performance
in terms of reliability, and developing an application that could help users manage the building.
6. References
[1] H. Gao, C. Koch, and Y. Wu, “Building information modelling based building energy
modelling: A review,” Applied Energy, vol. 238. 2019. doi: 10.1016/j.apenergy.2019.01.032.
[2] J. C. P. Cheng, W. Chen, K. Chen, and Q. Wang, “Data-driven predictive maintenance
planning framework for MEP components based on BIM and IoT using machine learning
algorithms,” Automation in Construction, vol. 112, 2020, doi: 10.1016/j.autcon.2020.103087.
[3] S. Tang, D. R. Shelden, C. M. Eastman, P. Pishdad-Bozorgi, and X. Gao, “A review of
building information modeling (BIM) and the internet of things (IoT) devices integration:
Present status and future trends,” Automation in Construction, vol. 101, May 2019, doi:
10.1016/j.autcon.2019.01.020.
[4] S. A. Khowaja, P. Khuwaja, K. Dev, and G. D’Aniello, “VIRFIM: an AI and Internet of
Medical Things-driven framework for healthcare using smart sensors,” Neural Computing and
Applications, 2021, doi: 10.1007/s00521-021-06434-4.
[5] F. Colace, C. Elia, C. G. Guida, A. Lorusso, F. Marongiu, and D. Santaniello, “An IoT-based
Framework to Protect Cultural Heritage Buildings,” 2021. doi:
10.1109/smartcomp52413.2021.00076.
[6] F. Colace, C. Elia, E. Landolfi, M. Lombardi, D. Santaniello, and A. Troiano, “An IoT-Based
Framework for Smart Agriculture Managing and Product Enhancing,” in Proceedings of Sixth
International Congress on Information and Communication Technology, 2022, pp. 111–119.
doi: 10.1007/978-981-16-2102-4_10.
[7] A. Castiglione, F. Colace, V. Moscato, and F. Palmieri, “CHIS: A big data infrastructure to
manage digital cultural items,” Future Generation Computer Systems, vol. 86, 2018, doi:
10.1016/j.future.2017.04.006.
[8] D. Capriglione et al., “Development of a test plan and a testbed for performance analysis of
MEMS-based IMUs under vibration conditions,” Measurement: Journal of the International
Measurement Confederation, vol. 158, 2020, doi: 10.1016/j.measurement.2020.107734.
[9] X. Gao and P. Pishdad-Bozorgi, “BIM-enabled facilities operation and maintenance: A
review,” Advanced Engineering Informatics, vol. 39. 2019. doi: 10.1016/j.aei.2019.01.005.
[10] C. Boje, A. Guerriero, S. Kubicki, and Y. Rezgui, “Towards a semantic Construction Digital
Twin: Directions for future research,” Automation in Construction, vol. 114. 2020. doi:
10.1016/j.autcon.2020.103179.
[11] S. H. Khajavi, N. H. Motlagh, A. Jaribion, L. C. Werner, and J. Holmstrom, “Digital Twin:
Vision, benefits, boundaries, and creation for buildings,” IEEE Access, vol. 7, 2019, doi:
10.1109/ACCESS.2019.2946515.
[12] T. Bloch and R. Sacks, “Comparing machine learning and rule-based inferencing for semantic
enrichment of BIM models,” Automation in Construction, vol. 91, 2018, doi:
10.1016/j.autcon.2018.03.018.
[13] I. Cvitić, D. Peraković, M. Periša, and B. Gupta, “Ensemble machine learning approach for
classification of IoT devices in smart home,” International Journal of Machine Learning and
Cybernetics, 2021, doi: 10.1007/s13042-020-01241-0.
[14] A. di Filippo, M. Lombardi, F. Marongiu, A. Lorusso, and D. Santaniello, “Generative design
for project optimization,” 2021. doi: 10.18293/dmsviva21-014.
[15] M. Henschke, X. Wei, and X. Zhang, “Data Visualization for Wireless Sensor Networks Using
ThingsBoard,” 2020. doi: 10.1109/WOCC48579.2020.9114929.
[16] AlZu’bi, S., Hawashin, B., Mujahed, M., Jararweh, Y., & Gupta, B. B. (2019). An efficient
employment of internet of multimedia things in smart and future agriculture. Multimedia Tools
and Applications, 78(20), 29581-29605.
[17] Al-Ayyoub, M., Al-Mnayyis, N., Alsmirat, M. A., Alawneh, K., Jararweh, Y., & Gupta, B. B.
(2018). SIFT based ROI extraction for lumbar disk herniation CAD system from MRI axial
scans. Journal of Ambient Intelligence and Humanized Computing, 1-9
[18] Al-Ayyoub, M., AlZu’bi, S., Jararweh, Y., Shehab, M. A., & Gupta, B. B. (2018). Accelerating
3D medical volume segmentation using GPUs. Multimedia Tools and Applications, 77(4),
4939-4958
[19] Al-Ayyoub, M., AlZu’bi, S., Jararweh, Y., Shehab, M. A., & Gupta, B. B. (2018). Accelerating
3D medical volume segmentation using GPUs. Multimedia Tools and Applications, 77(4),
4939-4958.
[20] AlZu’bi, S., Shehab, M., Al-Ayyoub, M., Jararweh, Y., & Gupta, B. (2020). Parallel
implementation for 3d medical volume fuzzy segmentation. Pattern Recognition Letters, 130,
312-318.
[21] Sandeep Kumar (2021) Artificial Intelligence and Machine learning for Smart and Secure
Healthcare System, Insights2Techinfo, pp.1
[22] A. Dahiya, K. Psannis (2021), Internet-of-Medical-Things (IoMT): An Unexplored Dimension
in Healthcare, Insights2Techinfo, pp.1
[23] Mamta (2021) Quick Medical Data Access Using Edge Computing, Insights2Techinfo, pp.1
[24] Ruchi Bhatnagar and Deepak Sinha (2021) Smart Adoption of IoT in Covid 19 Pandemic paves
new era of sustainable future world, Insights2Techinfo, pp.1
[25] A. Dahiya, B. B. Gupta (2021) How IoT is Making DDoS Attacks More Dangerous?,
Insights2Techinfo, pp.1
[26] Tripathi, S., Gupta, B., Almomani, A., Mishra, A., & Veluru, S. (2013). Hadoop based defense
solution to handle distributed denial of service (ddos) attacks. Journal of Information Security.
Vol. 4 No. 3 (2013) , Article ID: 34629 , 15 pages DOI:10.4236/jis.2013.43018