Automatic Test Data Generation Tool Based on Genetic Simulated Annealing Algorithm

This paper presents an automatic test data generation tool that aims at generating test data dynamically by reducing the problem of test data generation to one minimizing function. The tool is composed of three parts: program instrumentation module; test path generation module and test case generation module. In order to enhance the computational efficiency, we makes some improvements on the encoding, crossover probability and annealing gene of the genetic simulated annealing algorithm (GASS), and then use the results to generate the test data on the program instrumentation. Experimental results show this approach has better effect.