Behavior analysis and modeling of entities is discussed as one of practical methods for data oriented approach to systems design. For this purpose, the notions of normalization and abstraction of entities are introduced to refine conceptual Schemas in the framework of the ER(Entity-Relationship) model. Behavior of entities, which is defined as a set of states and transactions that bring about state transitions, is analyzed and modeled using the ER model and diagrammatic techniques of Petri Nets. A formalization of behavior description, together with the notion of transaction normalization, is also proposed. Based on the behavior modeling, an idea for data oriented approach to information systems design is discussed.

The purpose of this paper is to present a systematic approach to the conceptual schema design. The entity-relationship model is used as the conceptual schema model. The entity-relationship schema, a formal description of the model, is defined to explicitly state the dependency structures such as the functional dependencies, the transitive dependencies, and the hierarchical decompositions. Based upon the analysis of these structural properties, the schema is iteratively transformed into the refined form. The notion of the access weight matrix is also introduced as a measure to evaluate the schemas.