DEVELOPING A KNOWLEDGE DISCOVERY FRAMEWORK IN HIGHWAY PROJECTS USING INDUSTRIAL FOUNDATION CLASS (IFC) EXTENSION

1 online resource (135 pages) : PDF

2016

University of North Carolina at Charlotte

Highway-project data are being accumulated at a dramatic pace with the advance of information technology; consequently, the need for discovering useful knowledge among the historical data to prevent highway-project construction delay is increasing. However, the efficiency and effectiveness of most current highway-project knowledge discovery approaches are very low. A new systematic approach integrating Industrial Foundation Class (IFC), data warehouse, and data mining technologies is needed to address this issue. The proposed highway IFC data schema adopts an object oriented design approach to integrate interdisciplinary highway information into one universal intelligent format. IFC data is machine readable and suitable for complex query and reasoning, thus the highway-project knowledge discovery process can be highly automated by deploying IFC in the data warehouse with data mining function. This above-mentioned approach is tested with a 3D highway data warehouse prototype system in a case study of KICT project. The prototype system stores the highway IFC data and automatically generate the feature data and target data according to user requirements. Two data mining models are built to find useful patterns among the data. The performances of the models are compared. A cost sensitive analysis is especially performed to take into consideration of the practical cost of a project manager’s decision. Overall, this research has developed a highly automated approach to discover highway-project knowledge from historical data in an effective and efficient manner, which can help highway-project managers make timely and wise decisions to avoid potential construction delay and the associated costs.

doctoral dissertations

Computer science
Civil engineering

Ph.D.

BIM
Data Mining
Highway
IFC
KDD

Civil Engineering

Kim, Hyunjoo
Wu, Jy

Chen, Don
Lu, Na
Azarbayjani, Mona

Thesis (Ph.D.)--University of North Carolina at Charlotte, 2016.

This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s). For additional information, see http://rightsstatements.org/page/InC/1.0/.

Copyright is held by the author unless otherwise indicated.

Shen_uncc_0694D_11213

http://hdl.handle.net/20.500.13093/etd:611