Automated impact analysis of UML models

  • Authors:

  • L. C. Briand;Y. Labiche;L. O'Sullivan;M. M. Sówka

  • Affiliations:

  • Software Quality Engineering Laboratory, Systems and Computer Engineering Department, Carleton University, 1125 Colonel By drive, Ottawa, Ontario, Canada K1S5B6;Software Quality Engineering Laboratory, Systems and Computer Engineering Department, Carleton University, 1125 Colonel By drive, Ottawa, Ontario, Canada K1S5B6;Software Quality Engineering Laboratory, Systems and Computer Engineering Department, Carleton University, 1125 Colonel By drive, Ottawa, Ontario, Canada K1S5B6;Software Quality Engineering Laboratory, Systems and Computer Engineering Department, Carleton University, 1125 Colonel By drive, Ottawa, Ontario, Canada K1S5B6

  • Venue:
  • Journal of Systems and Software
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The use of Unified Modeling Language (UML) analysis/design models on large projects leads to a large number of interdependent UML diagrams. As software systems evolve, UML diagrams undergo changes that address error corrections and changed requirements. Those changes can in turn lead to subsequent changes to other elements in the UML diagrams. Impact analysis is defined as the process of identifying the potential consequences (side-effects) of a change, and estimating what needs to be modified to accomplish that change. In this article, we propose a UML model-based approach to impact analysis that can be applied before implementation of changes, thus allowing early decision-making and change planning. We first verify that the UML diagrams in a design model are consistent. Then the changes between two different versions of UML models are automatically identified according to a change taxonomy. Next, model elements which are directly or indirectly impacted by the changes (i.e., may undergo changes) are determined using formally defined impact analysis rules (defined with the Object Constraint Language). A measure of distance between a changed element and potentially impacted elements is also proposed to prioritize the results of impact analysis according to their likelihood of occurrence. We also present a prototype tool that provides automated support for our impact analysis strategy, and two case studies that validate both the methodology and the tool. Empirical results confirm that distance helps determine the likelihood of change in the code.