A change impact analysis approach using visualization method

A software needs to be maintained whenever new requirements arise. A maintainer is responsible to change the code to ensure the software is more operational based on the changes. It is a fact that the change in some parts of the code will affect other parts within the same code that need to be chang...

Full description

Saved in:
Bibliographic Details
Main Author: Mohamad, Rita Noremi
Format: Thesis
Language:English
Published: 2010
Subjects:
Online Access:http://eprints.utm.my/id/eprint/11505/6/RitaNoremiMohamadMFSKSM2010.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A software needs to be maintained whenever new requirements arise. A maintainer is responsible to change the code to ensure the software is more operational based on the changes. It is a fact that the change in some parts of the code will affect other parts within the same code that need to be changed and this requires the maintainer’s experience to understand the behavior of code implementation. However, it is difficult for new maintainers to understand the component dependencies within the code and it is worse when the current documentation is not updated and becomes obsolete. Hence, the only trustworthy source is the most recent version of the code that new maintainers need to master which requires familiarity of the software itself prior to any changes. This is unfortunately tedious and time consuming. Therefore, to reduce the burden, this research extended and enhanced Configuration Artifact Traceability Tool for Impact Analysis (CATIA) to improve understanding in visualization. A Change Impact Analysis (CIA) approach using visualization method is a proposed model to assist maintainers to understand and visualize the traceability of potential change impact. The visualization method was developed based on inheritance, friend, composition, calls and aggregation of program structure to support software change request and program understanding. A prototype tool called CIA-V was developed to support Object Oriented Programming written in C++ to implement potential change impact using visualization method. The model was then tested in controlled experiment using selected case study to prove traceability effectiveness of potential change impact via visualization method.