A knowledge-based Approach for the Assessment of Damages to Constructions
Al-Hakam Hamdan, Raimar J. Scherer
Abstract
To assess detected anomalies as well as assume undetected damages of an existing construction, a knowledge-based approach for damage evaluation is presented, whereby data from multiple web ontologies are linked via Linksets that are stored in an Information Container for Data Delivery (ICDD). In the ICDD, an ontological model that represents the assessed construction is linked with a damage ontology, which is necessary to identify various damage types that could affect the construction. Additionally, the web ontology representing the construction is linked with a Building Information Model (BIM) in the IFC-Format to enable access to the geometrical data of the components. Similarly, the damage objects are linked with either recorded geometry data or a manually created geometry model. Predefined rules are applied on the assertion components of the web ontology for reasoning damage types of each detected anomaly and damage assessment. Furthermore, the existence of undetected damages can be reasoned based on information about the construction, environment and previously classified damages. The developed web ontologies in this research use the principles of Linked Data and are serialized in the data format OWL. This enables the utilization of already existing Linked Data ontologies as well as supporting the implementation and development of future ones.