Exploring BIM Data by Graph-based Unsupervised Learning

Jin, Chaoyi; Xu, Minyang; Lin, Lan; Zhou, Xiangdong

doi:10.5220/0006715305820589

Cited by 11 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Graph representation of BIM is an emerging research area in the construction industry (Jin et al, 2018;Saad et al, 2023). It improves interoperability between different disciplines and facilitates accessing architectural, structural, and mechanical design knowledge from the BIM data.…”

Section: Graph Representation Of Bimmentioning

confidence: 99%

“…In other words, embedding involves generating a vector representation based on the features and attributes of a graph while attempting to preserve as much graph information as possible. Jin et al (Jin et al, 2018) present a graph-based unsupervised method to obtain functional knowledge from building space structures. They used the space properties and their boundary relationships for space clustering.…”

Section: Graph Neural Network (Gnn)mentioning

confidence: 99%

See 1 more Smart Citation

A graph-based explanatory model for room-based energy efficiency analysis based on BIM data

Kiavarz,

Jadidi,

Esmaili

2023

Front. Built Environ.

View full text Add to dashboard Cite

Introduction: In recent years, the growing interest in building energy consumption and estimation has led to a wealth of energy data and Building Information Modelling (BIM), providing ample opportunities for data-driven algorithms to be widely applied in the building industry. However, despite promising accuracy in data-driven models for building energy estimation, they only consider building elements and their attributes independently and neglect the interconnected relationship of building elements. Also, Current data-driven models lack interpretability and are often treated as black boxes. As a result, the models cannot be fully trusted for engineering without reasoning the underlying mechanisms behind the estimation.Method: This paper emphasizes the potential of graph-based learning algorithms, specifically GraphSAGE, in utilizing the enriched semantic, geometry, and room topology information derived from BIM data. The aim is to identify critical zones within the building based on their energy consumption characteristics. Besides that, the paper proposed a GraphSAGE explainable model by adopting the SHAP with the proposed NE-GraphSAGE prediction model to make more transparency behind the data-driven models.Results and Discussion: Preliminary results demonstrate the potential to improve pre-construction and post-construction steps by identifying critical zones in buildings and identifying the parameters which affected the efficiency of the zones with low energy consumption.

show abstract

Section: Graph Representation Of Bimmentioning

confidence: 99%

Section: Graph Neural Network (Gnn)mentioning

confidence: 99%

A graph-based explanatory model for room-based energy efficiency analysis based on BIM data

Kiavarz,

Jadidi,

Esmaili

2023

Front. Built Environ.

View full text Add to dashboard Cite

show abstract

“…A common tool for data exchange is the export of the model to IFC (Industry Foundation Classes) format. Recent research [26] has shown that 3D data from an IFC file can be processed using AI to be transformed into spatial contexts and relationships. Subsequently, together with data-driven workflows of energy performance of buildings, they would seek to improve the energy balance of buildings by increasing the accuracy of the calculation [27].…”

Section: Petrova Formulated the Main Goals And Outputs Of Her Research In Three Points [21]mentioning

confidence: 99%

Use of Artificial Intelligence in the Field of Sustainable Architecture: Current Knowledge

Krausková

Pifko

2021

Architecture Papers of the Faculty of Architecture and Design STU

View full text Add to dashboard Cite

Innovative technologies help automate the work of the designer. A 3D model of the building can be used to calculate the required values. This will also allow you to create associative sections that, when changing the geometry of the 3D model, automatically adjust the drawn elements on the resulting 2D drawing. Information technologies enable participants from all over the world to work on one project and, thanks to the BIM (building information modelling) method, to design buildings during their life cycle more efficiently. At present, critical studies are published on interoper-ability in BIM and its lack of coordination or amount of information that is misinterpreted, etc. However, working in BIM is the most effective way to use computer technology to design buildings. There is a lot of information about the building in the 3D model itself, which can also be used for purposes other than construction (building management, reconstruction). But how to process a large amount of information in a 3D model? Many buildings already have their 3D models shared on cloud platforms, these contain information that could help, for example, find solutions for green construction using artificial intelligence (AI). We meet with AI every day. It supports internet search engines, predicts auto-completion words as you type. AI can also be found in architecture – not only as visions at exhibitions, but also in research on process optimization in BIM.

show abstract

“…Data ordered with IFC can be reviewed and studied with BIM model checking software tools, such as Solibri. Then, it can be suitably mined manually, with semantic and/or latent techniques ( [53,54,55,56,57,58]), or with dedicated data parsers [59], and exported into file formats as input for ML suites; examples of such formats are .arff files for the Waikato Environment for Knowledge Analysis (WEKA), or structured .csv files to be incorporated in ML libraries of the Surprise Scikit, a Python-powered scientific toolkit for recommender systems. But for this data to be translated into meaningful independent input variables, and then connected with meaningful dependent output variables as part of a ML modelling (and especially SML) addressing the research gap mentioned in the previous section (namely, the absence of BIM data utilization for the prediction of a building project's performance, and especially its delivery cost and time overheads), it needs to be incorporated in a suitable theoretical and conceptual framework.…”

Section: Data In Ifcs and Constructability For Machine Learning Predimentioning

confidence: 99%

Building Information Models’ Data for Machine Learning Systems in Construction Management

Kifokeris¹,

Roupé²,

Johansson³

et al. 2019

Proceedings of the Creative Construction Conference 2019

View full text Add to dashboard Cite

Qualitative and quantitative data are important in construction management. However, despite the capabilities of construction informatics, such data and its sources have scarcely been fully and systematically utilized for predictive purposes. Building Information Models (BIM) are such a data source. Within BIM, information structures enabling interoperability and providing utilizable data throughout the various Levels of Development (LODs) of a buildingfor example, Industry Foundation Classes (IFCs)can be fully and meaningfully exploited through data mining, and more particularly, via machine learning. In this paper, the capabilities of the information structures found in IFCs to be used as data sources for developing machine learning predictive models, will be examined. In addition, and by conceptually tying such data with constructability, their suitability for predictingthrough such machine learning modelsthe delivery cost and time overheads of a construction project, will be considered.

show abstract

Exploring BIM Data by Graph-based Unsupervised Learning

Cited by 11 publications

References 4 publications

A graph-based explanatory model for room-based energy efficiency analysis based on BIM data

A graph-based explanatory model for room-based energy efficiency analysis based on BIM data

Use of Artificial Intelligence in the Field of Sustainable Architecture: Current Knowledge

Building Information Models’ Data for Machine Learning Systems in Construction Management

Contact Info

Product

Resources

About