Evaluating the reliability of the software of a switching system with a multi‐variable model
Patrizia Bondi, Guido Simonetti · Software Testing Verification and Reliability · 1995
Abstract In recent years reliability growth models have gained a dominant role in the evaluation of the reliability of software products. However, their application to the quality assurance process of the software of a highly complex system such as a telecommunications switching system often shows that it is then possible to get accurate results only when some months of operation have elapsed. It is then impossible to give an answer to the typical questions which arise during the quality assurance phase: ‘Is the software ready for release? How long will it take before it is ready?’. This paper describes a multi‐variable (MV) model which was developed to overcome the above limitation by providing a better model of the quality assurance process. A case study is presented, and a comparison is made between the results obtained by applying existing reliability growth models and the MV‐model to it. Besides attaining surprising accuracy in its predictions and in measuring the reliability of the product, the MV‐model also tries to tackle a different kind of question: ‘How is it possible to improve the effectiveness of testing?’.