An automated change impact analysis approach for User Requirements Notation models

Abstract

Requirements and their models often evolve to reflect changing needs, technologies, and regulations. The decision to implement proposed changes to requirements models relies on means to capture and analyze the potential impact of such changes. The User Requirements Notation (URN) is a standardized requirements modeling language that incorporates two complementary views based on Use Case Maps (UCM), for expressing scenarios and processes bound to architectural components, and on the Goal-oriented Requirement Language (GRL), for capturing the goals of actors and their relationships. This paper presents a new Change Impact Analysis (CIA) approach for URN models. Following a proposed change, this approach helps identify potentially impacted URN constructs within the selected GRL/UCM view, as well as throughout other view elements connected with URN links. This URN-oriented CIA approach is implemented as an extension of the jUCMNav modeling environment, and its applicability is demonstrated using an illustrative URN specification and three real and publicly available specifications. Furthermore, an empirical study involving 10 participants is used to assess the accuracy of this approach in identifying impacted URN elements upon specification changes. Results indicate excellent accuracy and a significant reduction in user-perceived difficulty when estimating the impact of changes in URN specifications.