Journal of Computing in Civil Engineering · 2017 · 112 citations · 46 references
ArchitectureCivil EngineeringConstruction ManagementBuilding Information ModelingConstruction EngineeringConstruction Project ManagementBim AuthoringBim For Facility ManagementBim Lifecycle ManagementBuilding ConstructionConstruction AutomationBim Data IntegrationBim-enabled Facility ManagementBuilding Information Model IntegrationBuilding Information Modeling (Bim) IntegrationBuilding Information Modeling AutomationBuilding Information Modeling Model UpdatingConstruction Information TransferIndustry Foundation Classes
Accurate and proactive inspection of constructed objects and identification of their defects and design conformity is vital to both facility management and construction projects. A project’s building information model (BIM) needs to be regularly updated to more accurately reflect the actual site conditions. However, conventional practices lack site-data integration with models and rely on manual, error-prone, and costly updates. This paper proposes and demonstrates a robust technique that uses the industry foundation classes (IFC) schema to automatically update an as-designed BIM based on site observations for inspected building elements. It receives as input an inspected object’s actual type and inspection details including the detected defects/changes, responsible actors, as-built/as-is type, captured images, and time and the date of the inspection. The algorithm automatically analyzes the IFC data model to retrieve the element’s semantics and identifies discrepancies between the as-built/as-is and as-designed object conditions. Based on the results of the discrepancy analysis, the IFC data model is populated with new semantics, resulting in the update of object types, properties, and three-dimensional (3D) shape representations. Inspection details and user entries are automatically documented in the BIM and assigned to objects to enable potential diagnostics and tractability. This automates site-to-BIM data transfer and supports reality-capture techniques.
46