A Model-Based Approach to Impact Analysis Using Model Differencing

Abstract

Impact analysis is concerned with the identification of
consequences of changes and is therefore an important
activity for software evolution.
In model-based software development, models are core artifacts,
which are often used to generate essential parts of a software system.
Changes to a model can thus substantially affect different artifacts of a software
system. In this paper, we propose a model-based approach to impact analysis, in which
explicit impact rules can be specified in a domain specific language
(DSL). These impact rules define consequences of designated UML class
diagram changes on software artifacts and the need of dependent activities
such as data evolution. The UML class diagram changes are identified automatically
using model differencing. The advantage of using explicit impact rules is that
they enable the formalization of knowledge about a product.
By explicitly defining this knowledge, it is possible to create a checklist
with hints about development steps that are (potentially) necessary to manage
the evolution. To validate the feasibility of our approach, we provide results
of a case study.