考虑发布后故障发现概率的软件费用模型  

Software Cost Model of Considering the Fault Detection Probability After Software's Publishment

作　　者：张烨[1] 吴开贵[1] 

机构地区：[1]重庆大学计算机学院,重庆400044

出　　处：《计算机系统应用》2012年第1期90-93,共4页Computer Systems & Applications

基　　金：国家自然科学基金(大型分布式软件系统的行为监控与可信演化项目编号:90818028)

摘　　要：在基于G-O模型的软件可靠性增长模型中引入以时间为变量故障察觉率,并以此模型建立了一种新的软件费用模型。该费用模型考虑了软件发布后使用者对软件系统剩余故障的发现概率,并不是所有的剩余故障全部被发现这种情况,使模型更符合实际。以软件开发费用最小为最优化条件,讨论发布时间与费用的关系,并在文章最后用示例说明了发布时间随参数变化。Leading time as a variable of fault detection rate into the G-O Model of Software Reliability Growth Model（SRGM）,we build a new Software Cost Model.This Software Cost Model considers the remaining fault detection probability of software system users after the software＇s publishment.It is a model that considers error is not completely found,so the model is more practical.The article is based on the least costs of the software as the best condition and talks about the relationship of publishing time and costs.At the end of this article,examples are showed to explain the publishing time changing by parameters.

关 键 词：非齐次泊松过程 故障察觉率 软件可靠性增长模型 软件费用模型 最优发布时间 

分 类 号：TP311.53[自动化与计算机技术—计算机软件与理论]