عنوان مقاله [English]
In software systems, the most of unexpected bugs usually occur when some of the system components interact with each other at the same time. Combinatorial testing is a method that aims to generate minimized test suite to determine the bugs caused by these components and their interactions. Covering array generation, an optimization problem, is the most popular research area in the field of combinatorial testing. Meta-heuristic algorithms have succeeded in generating covering arrays. Although good results can be found by these strategies, their complex search processes are time consuming. In this respect, these strategies have been confined to small configurations and they can support small interaction strengths (t<=6). In this research, teaching learning base optimization algorithm has been used and fitness function has been designed in a way that searching process become so fast that our strategy enabled to generate covering array with interaction strength up to t=15. In additition, ability of the algorithm in minimization of test suite is also high. The results of this current study showed that our proposed algorithm is able to generate the results more appropriate than the existing strategies.