Safety Critical Systems are those systems that can potentially lead to loss of life, injury, and environmental damage. Therefore such systems have to be designed and built to meet a variety of functional and non-functional requirements, including safety, reliability, availability, and maintainability. It is essential to assess, as an independent activity, the extent to which these requirements have been met, and for complex systems there is no single analysis technique which can be employed. It is therefore necessary to use a number of different safety (and reliability) analysis techniques to perform an assessment. Using a variety of techniques raises issues of consistency-if the individual analyses and models are inconsistent with respect to each other then the overall assessment is likely to be inconsistent, and therefore not trustworthy. In this paper we present a set of rules that should hold between a representative set of safety analysis techniques, demonstrate how they can be enforced and checked by an underpinning data model, and describe a software tool (based on these ideas) to support integraed safety analysis.
[1]
John A. McDermid,et al.
A development of hazard analysis to aid software design
,
1994,
Proceedings of COMPASS'94 - 1994 IEEE 9th Annual Conference on Computer Assurance.
[2]
Jim Woodcock,et al.
The Formal Specification in Z of Defence Standard 00-56
,
1994,
Z User Workshop.
[3]
P Holgate,et al.
Health and safety at work.
,
1975,
Nursing mirror and midwives journal.
[4]
W E Vesely,et al.
Fault Tree Handbook
,
1987
.
[5]
John A. McDermid,et al.
Support for safety cases and safety arguments using SAM
,
1994
.