KNOWLEDGE MANAGEMENT AND KNOWLEDGE REPRESENTATIONFOR SYSTEMS ANALYSIS

摘要

In this paper, the method for systems analysis basedrnon knowledge management is proposed. In the systemsrndevelopment, the system is analyzed in terms of severalrnsystem views to acquire the requirements sufficiently. Inrnmany cases, the system view is described by using chartrnrepresentation which consists of several kinds of nodesrnand arcs. The combinations of the nodes and the arcs arernaccumulated as knowledge and reused empirically. Torndescribe the knowledge, the authors propose KRSArn(Knowledge Representation for Systems Analysis) whichrnconsists of Topological Knowledge Primitives,rnGeometrical Knowledge Primitives and OperationalrnKnowledge Primitives. Topological KnowledgernPrimitives, Geometrical Knowledge Primitives andrnOperational Knowledge Primitives represent the logicalrnrelationships, the geometrical relationships and thernoperation respectively. In this paper, the authors proposernthe knowledge management using KRSA such as (a)rnacquisition of ontology for reactive system domain, (b)rnacquisition of ontology for trading task, ( c ) reuse of MCMrnfor collaboration system, (d) analysis process forrncollaboration task with many kinds of chart representationrnand (e) analysis process for collaboration task with PetrirnNet.