A New Approach to the Quantitative Measurement of Software Reliability

Nowadays software systems have very important role in a lot of sensitive and critical applications. Sometimes a small error in software could cause financial or even health loss in critical applications. So reliability assurance as a nun-functional requirement, is very vital. One of the key tasks to ensure error-free operation of the software, is to have a quantitative measurement of the software reliability. Software reliability engineering is defined as the quantitative study of the operational behavior of software systems with respect to user requirements concerning reliability. Software Reliability is defined as the probability of failure-free software operation for a specified period of time in a specified environment. Quantifying software reliability is increasingly becoming necessary. We have recently proposed a new approach (referred to as SDAFlex&Rel) to the development of «reliable yet flexible» software. In this paper, we first present the definitions of a set of key terms that are necessary to communicate with the scope and contributions of this work. Based on the fact that software reliability is directly proportional to the reliability of the development approach used, in this paper, a new approach is proposed to quantitatively measure the reliability of the software developed using SDAFlex&Rel, thereby making precise informal claims on the reliability improvement. The quantitative results confirm the reliability improvement that is informally promised by SDAFlex&Rel.