TY - GEN
T1 - Formal analysis and validation of levels of geometry (LOG) in building information models
AU - Abualdenien, Jimmy
AU - Borrmann, André
N1 - Publisher Copyright:
© EG-ICE 2020 Workshop on Intelligent Computing in Engineering, Proceedings. All rights reserved.
PY - 2020
Y1 - 2020
N2 - Construction projects are multidisciplinary and contractual. The collaboration among the project participants and the quality of the exchanged building information throughout the project lifecycle are prescribed in legal agreements. The Level of Development (LOD) concept is widely used for describing the building elements' maturity. Detailing models to a certain LOD is crucial for integrating the partial models as well as consumes additional time and costs. Every LOD comprises requirements for both Level of Geometry (LOG) and Level of Information (LOI). Thus far, the validation of LOD is limited to the LOI, whereas, checking the quality of the LOG is a complex and unsolved task. This paper proposes a framework for validating the LOG of building elements. In more detail, a LOG dataset is modelled, and then a formal metric is defined based on an extracted set of geometric features. Finally, a random forest model is developed for predicting the LOG.
AB - Construction projects are multidisciplinary and contractual. The collaboration among the project participants and the quality of the exchanged building information throughout the project lifecycle are prescribed in legal agreements. The Level of Development (LOD) concept is widely used for describing the building elements' maturity. Detailing models to a certain LOD is crucial for integrating the partial models as well as consumes additional time and costs. Every LOD comprises requirements for both Level of Geometry (LOG) and Level of Information (LOI). Thus far, the validation of LOD is limited to the LOI, whereas, checking the quality of the LOG is a complex and unsolved task. This paper proposes a framework for validating the LOG of building elements. In more detail, a LOG dataset is modelled, and then a formal metric is defined based on an extracted set of geometric features. Finally, a random forest model is developed for predicting the LOG.
UR - http://www.scopus.com/inward/record.url?scp=85091031655&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85091031655
T3 - EG-ICE 2020 Workshop on Intelligent Computing in Engineering, Proceedings
SP - 33
EP - 42
BT - EG-ICE 2020 Workshop on Intelligent Computing in Engineering, Proceedings
A2 - Ungureanu, Lucian-Constantin
A2 - Hartmann, Timo
PB - Universitatsverlag der TU Berlin
T2 - 27th EG-ICE International Workshop on Intelligent Computing in Engineering 2020
Y2 - 1 July 2020 through 4 July 2020
ER -