划分测试中测试用例最优分配问题研究

研究了划分测试中每个子域错误率已知，测试资源受约束时，测试用例如何合理分配的优化问题．主要考虑了两类测试用例分配问题：测试资源受约束的测试用例分配及测试资源受约束且各个子域的可靠性要达到一定要求时测试用例的分配，证明了带约束的测试用例最优分配问题是一个NPC问题．给出了动态规划求解方法，从理论上证明了该方法是最优的，并通过随机模拟将该方法与其它常见的测试用例分配策略进行分析比较，用实验数据验证了该办法能合理利用有限测试资源，有效地提高缺陷检测能力．

Optimal Allocation of Test Case Considering Testing-Resource in Partition Testing

We Investigated an optimal problem of determining the test allocation when the failure rate of each subdomain is given, and the testing-resource is constrained. We show that this problem can be formulated as two optimal models, maximization of the probability of detecting at least one fault when a fixed amount of cost objective is given and maximization of the probability of detecting at least one fault when a fixed amount of cost and a desired reliability objective are given. We transform it into a classic unbountied Knapsack problem to prove that it is a NPC problem; we present a dynamic programming algorithm that .solve the best allocation problem, and prove that it is the best result by dynamic programming algorithm. Our methodologies provide practical approaches to the optimization of testing-resource allocation with a reliability objective and a test cost objective. The experimental results by random simulation method greatly help us to identify that this method is consistently better than other test case allocation schemes. By using the proposed algorithms, project managers can allocate limited testing resource easily and efficiently and thus achieve the highest detecting fault ability objective during software testing.