Software reliability growth modeling based on testing-skill characteristics: Model and application

Abstract

In order to test a software system, many test-cases are executed in the final stage of software development. Then, the set of modules and functions are influenced by the executed test-cases. The set is called the testing-domain and it expands with the progress of testing. The growth rate of the testing-domain in the software system is closely related to the quality and quantity of executed test-cases. Further, the quality of test-cases is related to the testing skill of test-case designers. In this paper, we discuss a software reliability growth model which describes the time-dependent growth behavior of the isolated testing-domain. This model based on a nonhomogeneous Poisson process is used for describing a software fault detection phenomenon during the software testing. Finally, this model is applied to fault data observed in actual development projects, the software reliability assessment results are shown, and comparisons of goodness-of-fit with existing software reliability growth models are performed. © 2001 Scripta Technica, Electron Comm Jpn Pt 3, 84(6): 42–49, 2001