An analogy-based domain analysis methodology

Software reuse has the potential to increase productivity, improve quality, and reduce risk. Reusability is regarded as a main research subject in the area of software engineering. Domain analysis is a prerequisite for effective software reuse. Domain analysis is an expansion of the conventional requirements modeling. Domain analysis is defined as the process of identifying relevant information of a class of similar applications in a particular domain rather than just a specific system. However, domain analysis is suited to well-understood domains and is limited to a particular application area. Analogical approaches for software reuse, on the other hand, often occurs between different domains. Analogical problem solving is a process of transferring knowledge from a well-understood domain to a new problem area. Analogy can facilitate software reuse for poorly-understood problems or new application areas. But current research on software analogy overlooks the importance of domain analysis for the base domain and does not consider some critical aspects of software dynamics and analogy concepts reported in the literature. This dissertation presents an approach to software reuse, which integrates domain analysis and analogy methods. The approach is called analogy-based domain analysis (ABDA) methodology. In ABDA, domain analysis and software analogy play complementary roles. Domain analysis is regarded as a process to identify and supply necessary information for analogical mapping and transfer. Software analogy can provide the domain analyst with similar problems to start the domain analysis process. The dissertation emphasizes domain modeling techniques. A classification scheme for domain models is also presented. The methodology is then illustrated with case studies. Evaluation and classification of the methodology is conducted based on sets of criteria developed for requirements modeling, analogy, and domain analysis.