Software reliability measurement and assessment methods during operation phase and their comparisons
Shigeru Yamada, Shunji Osaki, Yoshikazu Tanio · Systems and Computers in Japan · 1992
Abstract This paper attempts to relate the testing phase, which is the final stage of the software development and the operation phase after the release. It discusses the occurrence of the software failure due to the errors latent in the software. The event of detecting the software error in the testing phase is described by the software reliability growth model based on a nonhomogeneous Poisson process. As to the assessment of the software reliability in the operation phase, i.e., the operational reliability, three assessment techniques are discussed. They are the method to introduce the environment factor considering the number of errors remaining at the end of the testing, the method based on the hazard rate, and the method based on the cumulative failure rate. Finally, the result of the analysis of the software reliability data is presented. Numerical examples of the assessment methods for the operational reliability are discussed and the results are compared.