软件可靠性关键参数判定的矩估计方法

Key parameter judgment with moment estimation in software reliability prediction

已有的基于结构的软件可靠性研究中大多假设建模参数为已知常量,导致模型忽略了参数估计的统计特征对于系统可靠性的影响.针对该问题,提出一种基于矩估计的软件可靠性关键参数判定方法,该方法通过分析基于结构软件可靠性模型的特点,建立各建模参数的矩估计量与系统可靠性预测结果之间的关系,量化各参数对于系统可靠性的影响,判定关键参数.为系统可靠性控制、系统结构优化及软件测试资源最优化分配等问题的解决提供了理论支持,通过对两个实际软件系统的分析验证了方法的有效性.

In most existed researches on architecture-based software reliability, parameters are mostly assumed as constants. But such an assumption is inconsistent with the practical situation for ignoring the impact on system reliability caused by statistical deviations of parameter estimation. For solving this problem, a judgment method for key parameters in software reliability analysis is proposed in this paper, which is based on the moment estimation theory. In this method, the characteristics of architecture-based software reliability models are analyzed, and the relationships between the system reliability prediction result and the moment estimation values of the parameters are given explicitly, then the impacts of the different parameters on system reliability can be analyzed quantitatively and the key elements can be judged simultaneously. The method can give some new theoretical supports for system reliability controlling, system architecture optimization and optimal allocation of software test resource. In the end of this paper, two actual examples of software system reliability analysis are given to illustrate the effectiveness of the method.