Research On BIM-based Contextual Information Retrieval Of Construction Engineering Document

There is huge amount information in construction engineering documents and most is unstructured information. Document information can hardly be fully utilized due to its operationality is limited and the lack of reusability and interoperability. At present, document information retrieval is realized by manual search or keyword-based search. Manual search is error-prone and time consuming. Keyword-based search also has some limitations due to the absence of semantic. They are often either too specific and exclude documents that would have been of interest, or they are too general and include too many documents, many of which are not of interest. Therefore, how to quickly extract the necessary information from the huge amount construction information especially from unstructured information is an important issue in current construction engineering document management.With the rapid popularization of BIM technologies, BIM integrates all project information and exchange the information by IFC. Through integrating the information in BIM and document can effectively solve the problem of document management in construction industry. Meanwhile, contextual information retrieval combines search technologies and knowledge about query and user context into a single framework in order to provide the most "appropriate" answer to meet user needs. An ontology-based model is one of the most popular context modeling methods. This paper describes the system of proposed BIM-based contextual information retrieval of construction engineering document.At first, this paper proposes an approach of automated develop IFC ontology. The differences of EXPRESS in IFC and OWL are compared. Through mapping converts the IFC format to OWL, the OWL can be input to protege5.0 to develop IFC ontology automatically. It’s an effective method for ontology-based context modeling.Secondly, this paper puts forward the process of Jena-based context reasoning. Predicate logic, premise and general reasoning rules are set to assist reasoning. Ontology-based reasoning and rule-based reasoning conducted in Jena API can extract the effective information which contained in the context to facilitate solve practical problems by making full use of the information.At last, a BIM-based construction engineering document contextual information retrieval system is proposed and the system’s feasibility and effectiveness are validated. JHA-IFC ontology is developed by protege based on JHA (Job Hazard Analysis) documents, reasoning rules are set to resaon by Jena API. The retrieval results show that the system can retrieve the context information related to the demand information. This system can greatly improve the effciency of construction safety management and the reusability of document information in construction industry.