故障察觉率可变的软件可靠性增长模型

摘要

证明了基于G-O模型的NHPP类型的软件可靠性增长模型不需要考虑不完美排错和排错过程中新错误的引入,并在该基础上提出了一种新的软件可靠性增长模型.该模型在软件排错过程中不但考虑了软件开发员对系统熟悉程度的上升,而且考虑了系统现存错误数的不断减少,是一种故障检测率随时间变化的软件可靠性增长模型.并利用现有的公开发表的数据对该模型进行测试,发现其达到了比G-O模型的等其他模型更好的拟合效果.%It is proved that SRGM based on NHPP type of G-O model doesn't need to consider imperfect debugging and new mistakes during the debugging process. A new type of SRGM comes out which does not only consider the software developer's familiarity with software system, but also consider the diminishing mistakes developing system. This SRGM has considered the fault detection rate changing with the time changing. Moreover, the test result by using the public reported data shows that the goodness of fit is better than that of other models of G-O model.