系统工程与电子技术

• 可靠性 • 上一篇    下一篇

考虑测试效用的软件可靠性增长模型

赵玉新,常帅,万程程   

  1. 哈尔滨工程大学自动化学院, 黑龙江 哈尔滨 150001
  • 出版日期:2014-09-25 发布日期:2010-01-03

Software reliability growth model based on test effort function

ZHAO Yu-xin,CHANG Shuai,WAN Cheng-cheng   

  1. College of Automation, Harbin Engineering University, Harbin 150001, China
  • Online:2014-09-25 Published:2010-01-03

摘要:

测试效用函数不仅可以描述测试所消耗资源随时间的变化,同时也是建立考虑测试效用的软件可靠性增长模型的前提和基础。现有测试效用函数形式多达十余种,但是由于适用性较差,导致研究人员对失效数据进行建模时需要付出繁琐的工作量,并且模型的评估能力往往达不到预期效果。针对该问题,提出了一种通用的测试效用函数形式,该函数形式更加灵活、适应能力更强。在此基础上,建立了基于该测试效用函数的软件可靠性增长模型。利用两组公开发表的数据进行验证,实验结果表明,该测试效用函数有效可行,考虑新的测试效用函数的软件可靠性增长模型具有更好的可靠性评估性能及适应能力。

Abstract:

The test effort function (TEF) can not only describe the change of resources consumed with time, but also be the premise and foundation of the software reliability growth model considering the test effort function. There are more than ten kinds of existing TEFs. Due to poor applicability, the researchers have very big workload in modeling and the result is not satisfactory. A new function form of TEF is proposed. The function form is more flexible and applicable. On this basis, the software reliability growth model based on the TEF is established. Two sets of public data are used to evaluate its effectiveness. The results show that the proposed function is effective and feasible and the 〖JP2〗software reliability growth model based on this function has better performance and adaptation.