City Research Online

Abstract

For safety-critical systems, the required reliability (or safety) is often extremely high. Assessing the system, to gain confidence that the requirement has been achieved, is correspondingly hard, particularly when the system depends critically upon extensive software. In practice, such an assessment is often carried out rather informally, taking account of many different types of evidence—experience of previous, similar systems; evidence of the efficacy of the development process; testing; expert judgement, etc. Ideally, the assessment would allow all such evidence to be combined into a final numerical measure of reliability in a scientifically rigorous way. In this paper we address one part of this problem: we present a means whereby our confidence in a new product can be augmented beyond what we would believe merely from testing that product, by using evidence of the high dependability in operation of previous products. We present some illustrative numerical results that seem to suggest that such experience of previous products, even when these have shown very high dependability in operational use, can improve our confidence in a new product only modestly.

Publication Type:	Conference or Workshop Item (Paper)
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Departments:	School of Science & Technology > Computer Science > Software Reliability

Export

Downloads