摘要
提出了一个软件可靠性分配模型 ,该模型在保证系统开发费用最低的前提下 ,将可靠性指标分配到各模块中 ;此外还运用了遗传算法来搜索模型的最优解 .实践表明 ,这是一个有效的途径 .
Software reliability allocation is an important subject of software engineering,which allocate certain reliability to all modules so that the system can reach the goal reliability.When designing software, engineers will estimate reliability of the software.If the software does not reach the goal, which modules should we choose to improve reliability and how much should we improve are what Software reliability allocation to handle. Software reliability allocation develops gradually recent years, however ,many reliability allocation methods allocate reliability to software without considering the cost of software. Some methods such as AHP considers the cost of software,but its algorithm is very complicated.In this article, we introduced a nonlinear programming model for reliability allocation. The model can minimize the cost of software, as well as guarantee that software system reliability can reach the goal. At the same time,we introduced genetic algorithm to achieve the optional result for the model,it proves to be a valid way.How to estimate model parameters is very important for a model,we introduce some methods to decide model parameters.To make our model more pratical,we need continue to make a good number of experiments.
出处
《武汉大学学报(理学版)》
CAS
CSCD
北大核心
2003年第1期44-48,共5页
Journal of Wuhan University:Natural Science Edition
基金
国家自然科学基金资助项目 (60 1 730 63)