萤火虫算法在测试用例集约简中的应用  

Application of firefly algorithm in test suite reduction

作　　者：宫云战[1] 徐健豪 邢颖[2] GONG Yunzhan;XU Jianhao;XING Ying(State Key Laboratory of Networking and Switching Technology,Beijing University of Posts and Telecommunications,Beijing 100876,China;School of Automation,Beijing University of Posts and Telecommunications,Beijing 100876,China)

机构地区：[1]北京邮电大学网络与交换技术国家重点实验室,北京100876 [2]北京邮电大学自动化学院,北京100876

出　　处：《哈尔滨工程大学学报》2020年第4期577-582,共6页Journal of Harbin Engineering University

基　　金：国家自然科学基金项目(61702044);中央高校基本科研业务费专项资金项目(2019XD-A20).

摘　　要：为了研究智能算法在测试用例集约简中的应用,本文提出一种基于萤火虫算法的测试用例集约简算法。将测试用例集约减问题转化为二元优化的集覆盖问题;对萤火虫算法进行改进,通过设置萤火虫细胞的方法,将萤火虫位置转化为0/1序列;将萤火虫算法与贪心算法相结合,进行求解。对不同程序和不同规模的测试用例集的仿真实验,在相同的测试次数下,萤火虫算法得到最优解的次数高于贪心算法和优化的GRE算法,萤火虫算法约简后测试开销比贪心算法减少了12.6%,比优化的GRE算法减少了2.3%。结果表明:该算法在约简能力和稳定性方面均优于传统算法。To study the application of an intelligent algorithm to test suite reduction, in this paper, we propose a test-suite-reduction algorithm based on the firefly algorithm. First, we transform the test-suite-reduction problem into a binary-optimization-set-coverage problem. Next, we improve the firefly algorithm by establishing firefly cells and convert the cell positions into a 0/1 sequence. Lastly, we combine the firefly algorithm with the greedy algorithm to obtain the best solution. The test-suite simulation results using different programs and scales show that, with the same number of tests, the firefly algorithm can obtain the optimal solution more often than both the greedy algorithm and the optimized GRE algorithm. The cost associated with the firefly algorithm is 12.6% lower than that of the greedy algorithm and 2.3% lower than that of the optimized GRE algorithm, which indicates that this algorithm is superior to traditional algorithms in terms of its reduction ability and stability.

关 键 词：软件测试 回归测试 测试用例集约简 二元优化 智能算法 萤火虫算法 集覆盖问题 分支覆盖 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]