An Automated Test Case Generation Approach by Genetic Simulated Annealing Algorithm

Abstract

A new automatic test data generation approach is presented based on length_N coverage criterion through genetic simulated annealing algorithm. The method mainly works on test data generation where the problem of test data generation is reduced to one of minimizing a function. Firstly, a new test coverage criterion called length_N is proposed to aim at the feasible of path coverage criteria. To enhance the generated efficiency, we make some improvements on genetic simulated annealing algorithm and use the results to generate test data. Lastly, experiment results show the method has a better effect.