Towards a better control of change impact propagation

Abstract

The software evolution requires an exhaustive understanding of all the artifacts or components composing the software. In fact, any intended change may generate an impact propagated to the various software components through the different kind of relationships relating them. This situation is generally called the ripple effect and may cause several types of damages or side effects. In this paper, we focus on the potential flow of change impact propagation as provoked by the carried out change. For that purpose, we propose a structural model of software evolution (SMSE). This model is intended to represent and manage the different software artifacts and their relationships. The instantiated SMSE is used to built a knowledge base system providing, among other facilities, an assistance to the designer (or engineer) of software evolution for an a priori assessment of the change impact. The major aim of the system is to help a systematic analysis of the change impact and the identification of its propagation.