基于改进蚁群算法UML状态机图的测试用例生成

Test cases generation from UML state machine diagrams using improved ACO

为了减少测试工作量,防止测试用例数目过多而导致爆炸,提出了使用改进的蚁群算法直接从UML状态机图中生成测试用例的方法。利用UML状态图是对系统可能的生命历程进行建模,由于状态图具有直观性,所以简化了生成测试用例过程,从而减少了测试的工作量。通过构造人工蚁群以及动态更新信息素矩阵,让蚂蚁选择还没有搜索或边信息素最小的顶点,确保了测试用例的数量既满足测试需求,又不会过多。通过编写算法及模拟仿真,对该算法的可行性和有效性做了验证和仿真分析。

To cut down the amount of test working and avoid to number exploding of test cases,a method that using improved ant colony optimization（ACO） generate test cases directly from UML state machine diagrams is presented.Firstly,it models the path of the potential life of the system utilizing the UML state machine diagram.Because the state diagram is visual,the method simplifies the process of test cases generation.Therefore,the method reduces the amount of test working.Then,the ants will select the pheromones which is the smallest or the vertex which is not funded through constructing the artificial ants and dynamic updating the pheromone.So the method guarantee that the number of test cases is either satisfied the requirement of testing or not exceeding.Through the preparation of algorithms and simulation,they validate and analysis the feasibility and effectiveness of the algorithm.