OMBMDID: A Preliminary Attempt at Automatic Utilization of Knowledge Contained in Mechanical Design Documents

摘要

Studies show that, in engineering design, above 70% of designer's working time is spent on the searching of design documents and the execution of simple but repeating design tasks like looking up table, calculation, simple reasoning, etc. Can we develop a kind of intelligent design document to make these simple tasks done by computers instead of designers? For this purpose, by borrowing idea from the Semantic Web vision and employing ontology and multi-agent system (MAS) technologies, an architecture of Ontology- and MAS-Based Mechanical Design Intelligent Documents (OMBMDID), which can enable intelligent agents to locate needed document segments oriented by their content, grasp the knowledge contained in documents and put it into use automatically, is proposed. First, similar works are reviewed. Then, aiming at knowledge contained in mechanical design documents, semantic annotation granularity and semantic annotation categories are analyzed. On the basis of above works, the architecture of OMBMDID is proposed. In the end, by studying a case about a piece of shaft design document, the feasibility of this architecture is proved.