A hybrid optimization algorithm for pairwise test suite generation

Pairwise testing is an effective combinatorial test generation technique that can generate relative small test suite to cover all pairs of parameter values at least once. Genetic algorithm has been used for pairwise test suite generation by some researchers. In order to improve the performance of genetic algorithm, this paper proposes a hybrid optimization algorithm by augmenting genetic algorithm with two-stage hill climbing. The first stage is to improve all the individuals after genetic operations. The second stage is to improve the best solution of the current generation at the end of each generation. A series of experiments are conducted to evaluate the proposed algorithm. Experiment results show that the proposed algorithm is very competitive with respect to other approaches reported in the literature.