A rule-based approach for the identification of quality improvement opportunities in GRL models

Goal-oriented modeling languages have been proposed to elicit, analyze, and document high-level system requirements in the early stages of the requirements engineering (RE) process. Problems during this stage may disseminate to the subsequent stages in the software development process and artifacts. Therefore, improving the quality of goal models would improve the quality of the requirements and, consequently, the quality of the developed system. This paper proposes an approach to help modelers identify quality improvement opportunities in Goal-oriented Requirements Language (GRL) goal models. To this end, a list of GRL bad smells (i.e., bad quality symptoms) is introduced and evaluated by experts. Then, an automated rule-based technique is proposed to detect the instances of these smells. The proposed approach is evaluated using a dataset gathered from academic and real-world projects. The results show that the developed technique could successfully detect the instances of the proposed bad smells in the evaluation models. We also found that the instances of the proposed bad smells were prevalent in both academic and industrial settings. The proposed bad smells and the detection technique provide a tool to locate quality improvement opportunities in GRL goal models.