基于改进AHP-Fuzzy的面向测试的软件质量评价模型  

A Software Quality Evaluation Model for Software Testing Based on Improved AHP-Fuzzy

作　　者：王勇利[1] 王鑫 Wang Yongli;Wang Xin(No.91404 Unit of PLA,Qinhuangdao 066001,China)

机构地区：[1]中国人民解放军91404部队,河北秦皇岛066001

出　　处：《计算机测量与控制》2020年第3期260-266,271,共8页Computer Measurement &Control

摘　　要：通用软件质量评价模型过于庞大不便于其在软件测试活动中直接应用,软件测试评价空间有限不足以消除传统AHP等方法中人为主观影响,以上诸多因素限制了面向测试的软件质量评价在工程实践中的有效开展;鉴于此,为解决面向测试的软件质量评价难以有效开展的问题,基于对传统AHP-Fuzzy模型的适当改进,提出一种面向测试的软件质量评价模型--SFSE(system failed state evaluating model,系统失效评估模型);SFSE模型以软件测试活动实测数据为输入,首先对各叶子节点进行单因素实测评价,之后采用从下至上、逐层迭代的方式对各级评估原子集实施评价,最终实现对被测软件质量状态的整体评估;实验结果表明,拓展综合评价集向量提供了可调节的量化评估参数,利用基于实测问题规模构建评估模糊评价矩阵、将权重投射至评价因素集上的模糊子集等方式,SFSE模型可在无需人工干预的情况下实现对被测软件质量状态的定量评价。To solve the problem that the software quality evaluation for software-testing cannot be effective in engineering practice,combining the characteristics of software testing implementation,SFSE Model(system failed state evaluating model),a new kind of software quality evaluation model for software testing,is proposed.The SFSE Model,well adapted for the characteristics of software testing implementation,is based on the appropriate modification of the traditional AHP-Fuzzy model.On the basis of proposing concepts such as System Failed State Evaluating,Synthetic Evaluating Set and so on,the SFSE Mode evaluates all leaf node based on the information of software-testing executing firstly.Then the SFSE Mode implements evaluation for every Atomy-Evaluating-Scope at all levels in the form of iteration step by step.The experimental results show that the Expanded-Synthetic-Evaluating-Set provides a quantitative assessment of the adjustable parameters and an adjustable mechanism for quantitative evaluating,based on the measured problem size building assessment fuzzy evaluation matrix and a weighted projection to the evaluation factor set of fuzzy subset,and the SFSE Model can automatically implement each node and whole software quality evaluation in the form of without human intervention.

关 键 词：软件测试 软件质量评价 层次分析 模糊综合评价 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]