摘要
为了有效利用资源、优化测试数据,在软件可靠性测试中引入试验设计:在操作剖面开发过程中应用正交试验设计,选择典型操作;在测试数据生成过程中使用组合设计,结合发生概率得到优化测试数据集.完整的软件可靠性测试过程包括确定系统模式剖面,然后根据正交试验设计确定操作剖面,根据组合设计确定运行分类,最后生成可靠性测试数据执行测试.在此基础上开发了测试数据自动生成工具,进行了实例应用.结果表明该方法不仅能为软件生成可靠性测试数据,还能发现更多的软件缺陷.
Experimental design was introduced to software reliability test to take full advantage of resources and optimize test data. Typical operations were selected by orthogonal experiment design in the development of operational profiles and optimized results were produced by combinatorial design in the generation of test data considering of occurrence probabilities. A full software reliability test consists of determining the system modes profile, developing operational profiles by orthogonal experiment design, choosing run categories by combinatorial design and generating reliability test data to execute test finally. An automation test data generation tool was developed according to the method and an example was applied with it. Results show that it can not only generate software reliability test data but also finding more software defects.
出处
《北京航空航天大学学报》
EI
CAS
CSCD
北大核心
2008年第12期1379-1383,共5页
Journal of Beijing University of Aeronautics and Astronautics
关键词
可靠性
测试
概率
缺陷
reliability
testing
probability
defects