Domain analysis is essential to core assets development in software product line engineering. Most existing approaches, however, depend on domain experts’ experience to analyze the commonality and variability of syst...Domain analysis is essential to core assets development in software product line engineering. Most existing approaches, however, depend on domain experts’ experience to analyze the commonality and variability of systems in a domain, which remains a manual and intensive process. This paper addresses the issue by proposing a model-driven approach to automating the domain requirements derivation process. The paper focuses on the match between the use cases of existing individual products and the domain functional requirements of a product line. By introducing a set of linguistic description dimensions to differentiate the sub-variations in a use case, the use case template is extended to model variability. To this end, a transformation process is formulated to sustain and deduce the information in use cases, and to match it to domain functional requirements. This paper also presents a prototype which implements the derivation as a model transformation described in a graphical model transformation language MOLA. This approach complements existing domain analysis techniques with less manual operation cost and more efficiency by automating the domain functional requirements development.展开更多
文摘Domain analysis is essential to core assets development in software product line engineering. Most existing approaches, however, depend on domain experts’ experience to analyze the commonality and variability of systems in a domain, which remains a manual and intensive process. This paper addresses the issue by proposing a model-driven approach to automating the domain requirements derivation process. The paper focuses on the match between the use cases of existing individual products and the domain functional requirements of a product line. By introducing a set of linguistic description dimensions to differentiate the sub-variations in a use case, the use case template is extended to model variability. To this end, a transformation process is formulated to sustain and deduce the information in use cases, and to match it to domain functional requirements. This paper also presents a prototype which implements the derivation as a model transformation described in a graphical model transformation language MOLA. This approach complements existing domain analysis techniques with less manual operation cost and more efficiency by automating the domain functional requirements development.
