Optimization of Test Data Generation Using Genetic Algorithm for Software Testing

Abstract

Software testing is an essential phase of the software development life cycle that ensures the reliability, correctness, and quality of software applications. However, the manual generation of effective test data is time-consuming, labour-intensive, and often unable to provide sufficient code coverage for complex software systems. Automated test data generation can address these limitations by systematically identifying input values capable of exercising different execution paths and detecting software faults. This study proposes a Genetic Algorithm (GA)-based optimization approach for automated test data generation, in which candidate test data are represented as chromosomes and evolved through selection, crossover, and mutation operations. A fitness function is formulated using testing objectives such as code coverage, execution path coverage, and fault-detection capability to identify highly effective test data. The proposed approach iteratively improves the quality of generated test cases while reducing redundant and ineffective inputs. The performance of the GA-based method can be evaluated using parameters such as statement coverage, branch coverage, path coverage, number of faults detected, test-suite size, and execution time. Experimental analysis is intended to demonstrate that evolutionary optimization can generate more effective and diverse test data than conventional or randomly generated test inputs. The proposed methodology provides an automated and scalable framework for improving software testing efficiency and reducing the effort required for comprehensive test-data generation. The approach can be further extended to multi-objective optimization and hybrid intelligent techniques for testing large-scale and complex software systems.