Machine Learning Based Approach to Building Element Classification for Semantic Integrity Checking of Building Information Models

2018 ◽  
Vol 23 (4) ◽  
pp. 373-383
Author(s):  
Bonsang Koo ◽  
Youngsu Yu ◽  
Raekyu Jung
Author(s):  
Okuhle Vonco ◽  
Jan Wium

<p>The paper describes a risk-based approach to enable construction teams to predict potential areas of rework. This is achieved by capturing historic construction data of concrete elements using Building Information Models (BIM), augmented by manual capturing by project parameters.</p><p>The approach consists of two parts. In the first part data is captured of relevant project parameters that may impact on rework. This data is stored in a database and relationships are determined between these factors and the occurrence of rework using a machine learning approach. In a second part concrete elements in a BIM is verified against the database to determine the rework risk of the element.</p><p>The approach will enable construction teams to pro-actively manage the construction process to reduce the probability of rework with resulting savings in time and cost.</p>


Author(s):  
Behnam Atazadeh ◽  
Leila Halalkhor Mirkalaei ◽  
Hamed Olfat ◽  
Abbas Rajabifard ◽  
Davood Shojaei

Sign in / Sign up

Export Citation Format

Share Document