Composing Frankensteins:Data-Driven Design Assemblies through Graph-Based Deep Neural Networks

As, Imdat; Pal, Siddharth; Basu, Prithwish

doi:10.35483/acsa.am.107.90

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even in the same data format, the requirements for extracting information are different in different usage scenarios, which further hinders the implementation of automatic conversion. For example, given a BIM model, BIM semantic enrichment requires triangular meshes (Collins et al, 2021), while floor plans evaluation requires a bubble diagram (As et al, 2019).…”

Section: Challengesmentioning

confidence: 99%

“…Generative design. GNNs can learn not only explicit design rules but also implicit design rules from existing designs, enabling more complex alternative designs (As et al, 2019). Because GNNs are an end-to-end learning model, raw data can be directly fed into the model without manual feature extraction, and the design rules can be learnt by GNNs automatically, reducing the requirement for programming skills.…”

Section: Opportunities Knowledge Graphmentioning

confidence: 99%

“…Compared with the classic generative design, a fully trained GNN-based generative design can directly generate suitable designs without iteration and evaluation, thus reducing the computing costs. As stated in the previous section, the current adoption of GNNs in generative design is limited in the generation of floor plans (As et al, 2019;Nauata et al, 2020;Chang et al, 2021). The potential of GNNs in generative design has not been fully explored.…”

Section: Opportunities Knowledge Graphmentioning

confidence: 99%

See 2 more Smart Citations

Graph neural networks in building life cycle: a review

Yilong¹,

Wang²,

Reza³

et al. 2022

Proceedings of the 2022 European Conference on Computing in Construction

View full text Add to dashboard Cite

Graph neural networks (GNNs) have attracted much attention in the field of machine learning because of their excellent performance on graph data. Graph data in the architecture, engineering and construction (AEC) sector is very common, such as bubble diagrams for space planning and point clouds for scan-to-BIM models. Some studies in AEC have adopted GNNs to solve practical problems. However, there has been a limited focus on the outcomes of these studies. Therefore, this paper aims to review the applications of GNNs in the building lifecycle. A wide range of existing literature was retrieved. The result shows that the adoption of GNNs is still in its infancy but has been increasing dramatically in recent years. Ten application domains were identified from the planning stage to the operation stage. In addition, the challenges and opportunities of GNNs adoption in AEC were discussed providing directions for future research.

show abstract

Section: Challengesmentioning

confidence: 99%

Section: Opportunities Knowledge Graphmentioning

confidence: 99%