ISSN 2996-8223
International Journal of Computer Science and Technology | Vol. 7, No. 3, March 2005 | pp. 17–24
DOI: 10.46882/2005/IJCST/070306
Research Article
Title: An Evolutionary Approach to Automated Software Test Case Generation Using Genetic Algorithms
Names of Authors: Hans-Werner Schmidt¹, Clara Mendonça²
Authors’ Affiliations: ¹Institut für Informatik, Technische Universität München, Munich, Germany; ²Departamento de Informática, Universidade de São Paulo, São Paulo, Brazil
Abstract:
Manual generation of comprehensive test suites for software validation is often error-prone and consumes significant developer hours. This paper details an automated framework that leverages Genetic Algorithms (GAs) to synthesize structurally optimal test cases based on branch coverage criteria. The fitness function balances control flow graph complexity with path coverage milestones to guide the evolutionary selection process. The system was benchmarked against five open-source Java projects containing a cumulative total of more than 15,000 lines of code. The genetic engine reached a comprehensive branch coverage metric of 94.7% within an execution window of G = 120 generations, a significant improvement over random generation techniques which plateaued at 61.2%. The rate of redundant test case generation was reduced to less than 4.5%, directly lowering the computational load needed for full software regression testing passes.
Keywords: Automated Software Testing, Genetic Algorithms, Branch Coverage, Evolutionary Computing, Test Case Generation
Manuscript Timeline: Received December 08, 2004; Revised January 22, 2005; Accepted February 15, 2005; Published March 25, 2005
Citation: Schmidt, H. W., & Mendonça, C. (2005). An Evolutionary Approach to Automated Software Test Case Generation Using Genetic Algorithms. International Journal of Computer Science and Technology, 7(3), 17–24.
Subscribe to read the full article: https://www.internationalscholarsjournals.org/subscribe-to-read