A fast offset phase calibration method

Gu, Xuemai; Hao, Xiaolin; Li, Xinxing; Guo, Song

doi:10.1109/icocn.2017.8121321

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned above, this concept should support organisations that work with BIM data from many different sources. Among them, there are different needs for the accuracy of the extracted information, while the results of machine learning algorithms always have to deal with a certain degree of uncertainty [32]. These uncertainties are of different relevance to stakeholders in a project.…”

Section: Proof Of Conceptmentioning

confidence: 99%

Concept for ai‐supported information allocation based on IFC data

Faltin,

Gille,

Jäkel

2023

ce papers

View full text Add to dashboard Cite

Across the different phases of a construction project, many actors are involved in the design, the construction, and the operation of the structure. Each actor and all stakeholders in the project require different information structures. Unifying these information structures has great benefits for the recipients of the information. This way, inspection routines, certifications or approvals could be automated. Within the framework of the BIM methodology, there are already solutions like the Level of Information Need (LOIN) or attribute mapping via translation tables. But the required effort to tailor the information for all stakeholders is considerable. Therefore actors or stakeholders who often receive IFC data from different sources with different structures face difficulties to automate their processes. In this paper, a concept is presented that uses deep learning techniques to translate between specific information structures of different stakeholders. In this way, the effort of providing information can be reduced. The proof of concept shows that a LSTM model can extract the desired information from an IFC file. It is also weighed up how the concept is to be evaluated, what opportunities and challenges the application may face.

show abstract

Section: Proof Of Conceptmentioning

confidence: 99%