Assessing quality of use case specifications

Abstract

Use case modeling is one of the popular documentation strategies meant to describe the functional requirements of a software system. However, due to the use of some Natural Language (NL) text issues, such as ambiguity, incorrectness, inconsistency, and incompleteness are inherently introduced in the specification, which can cause a significant impact on the quality of software under development. Typically, use cases are widely used as input in various software development phases such, as requirement analysis, domain analysis modeling, test case generation, planning and estimation, and maintenance. Hence, the quality of use cases is extremely important in practice. In this thesis work, the quest has been to identify the set of aspects which play a major role in assessing the quality of use case specifications (UCSs). For this, we aimed to achieve the following objectives: 1. Identifying the set of aspects from the existing literature to assess the quality of requirements specification and use cases. 2. Formulating the set of use case metrics used to assess the quality of UCSs. 3. Conducting two separate experimental studies, a replication of one another, to validate the correctness of formulated use case metrics. 4. Presenting a tool-support to facilitate automated computation of use case metrics Our findings of the thesis revealed that the proposed (or formulated) set of use case metrics are useful in assessing the quality of UCSs and the developed tool-support highlighting the specification defects and related issues for improvements.