An automated methodology for the design of usecase package diagram

The Software Requirements specification (SRS) of an information system contains description of various usecases that can be clustered as real worlds for different actors. Though, usecase diagram depicts the real world for each actor, there may be presence of some secondary actors for each real world. Thus, the real worlds of the actors are not completely isolated real worlds; there exists some sort of overlapping. This overlap indicates some sort of redundancy in the design diagrams of the same type. When this is transformed into code, the redundancy persists in the code also. The redundancy in the code always gives scope for inconsistency and defers integrity. To avoid this, there is a necessity to identify the usecase hierarchies and reflect them in usecase package diagram. This paper attempts to use the usecase diagrams, then develops automated methodology to identify the object classes participating in each usecase. It also proposes an automated methodology for design of usecase hierarchies based on "Extends", "Uses" and Groups". This paper considers the SRS of an information system. Then, using [1], molds it, abstracts its control flow in the form of CFT [1, 2, 3], designs DFT [1, 2, 3], of it & then abstracts the relevant sequence of statements for an each usecase of an actor. This paper presents our proposed methodology that identifies all possible interrelationships between usecases, forms the hierarchies of usecases using de-facto standards of hierarchical levels [4, 5]. This reorganization eliminates the redundancies and packages the usecase incorporating the common reuse and common closure principles. Here, we have also proposed the use of class diagram for the design of usecase hierarchies which may later used for the development of usecase package diagram.