Static slicing of Use Case Maps requirements models

Abstract

Requirements specification is a crucial stage in many software development life cycles. As requirements specifications evolve, they often become more complex. The development of methods to assist the comprehension and maintenance of requirements specifications has gained much attention in the past 20 years. However, there is much room for improvement for model-based specifications. The Use Case Maps (UCM) language, part of the ITU-T Z.151 User Requirements Notation standard, is a visual modeling notation that aims to describe requirements at a high level of abstraction. A UCM specification is used to integrate and capture both functional aspects (based on causal scenarios representing behavior) and architectural aspects (actors and system components responsible for scenario activities). As UCM models evolve and grow, they rapidly become hard to understand and to maintain. In this paper, we propose a static slicing technique to enhance the comprehension of UCM models. The developed slicing approach is implemented within the jUCMNav tool. We validate the proposed approach using a mock system and three publicly available UCM specifications. The results suggest that, on average, the models can be reduced by about 70% through slicing without losing information required for comprehension and maintenance activities. A small experiment involving 9 participants also suggests that the understandability of UCM specifications and comprehension speed have both improved substantially by using jUCMNav’s new slicing feature.