A testing‐domain‐dependent software reliability growth model for imperfect debugging environment and its evaluation of goodness‐of‐fit

Abstract

SUMMARYMany software reliability growth models which reflect various testing environments in the software development and operation/maintenance environment as a method for attaining a highly reliable software system have been proposed. These models assume a perfect debugging environment. That is, the detected faults are assumed to be corrected and removed certainly without introducing new faults. If we describe the software fault-detection phenomenon more exactly, it is realistic to assume that new faults can be introduced by the debugging activities. In this paper, we propose a software reliability growth model based on the testing domain in the software system, which is isolated by the executed test cases in software testing. In particular, our model assumes an imperfect debugging environment in which new faults are introduced in the fault-correction process, and is formulated as a nonhomogeneous Poisson process. Further, it is applied to fault-detection data observed in actual development projects, the results of software reliability assessment are shown, and comparisons of goodness-of-fit with the existing software reliability growth model are performed.

A replicated empirical study of a selection method for software reliability growth models

“…For example, in general, all software reliability models assume that the defect detection that occurs during testing follows an operational profile (Lyu 1996;Musa 1999). In addition, with some exceptions (e.g., Fujiwara and Yamada 2003), most reliability models assume a perfect debugging environment, i.e., the defects are assumed to be corrected immediately without inserting any new faults. In practice, software faults may not always be fixed during debugging, and new faults may be introduced.…”