划分测试用例选择的风险决策方法

Risk decision approach to test allocation problem in partition testing

研究了划分测试中划分方案确定情形下,如何在各个子域中合理选择测试用例问题。在充分利用测试者对待测软件导致错误输入分布信息了解的基础上,提出了一种基于风险决策的测试用例选择模型,给出测试用例选择的期望风险准则,并证明了该方法选择的测试用例集具有较高的缺陷检测能力。最后采用蒙特卡罗方法进行仿真模拟,结果表明该方法明显优于划分测试用例的其他选择策略。

This paper investigated the problem of determining the test allocation when a particular partitioning scheme had been chosen. This paper showed that this problem could be formulated as a risk decision problem of decision-making under un-certainty,and proposed a expected risk criteria to resolve this kind of problem. Presented an algorithm that solve the test allocation problem based on this criterion,and evaluated this criterion by means of a simulation experiment. The experimental results by Monte Carlo method greatly help us to identify this method is consistently better than the other test case allocation criteria.