Reuse in dependability cases
reuse safety arguments, structure, hazard analysis
Descriptive dependability arguments have become a standard part of the process of determining the dependability of a system. At the centre of this demonstration process is the use of techniques for systematic hazard analysis. Hazard identification, classification and mitigation techniques establish that either hazards can be avoided or that they will not affect the dependability of the system. To aid this process, descriptive arguments are commonly produced to mitigate the perceived severity of hazards. In such a process there are two main requirements that need to be fulfilled, that the analysis has (1) sufficient rigour and (2) sufficient coverage. Confidence in the rigour of a safety case, of which hazard analysis is a component, is directly linked to confidence in the hazard analysis itself.
As a step towards understanding confidence in coverage a number of overall measures of reuse of an argument were considered. As a result of this consideration the extent of reuse of arguments, including verbatim reuse of structure was analysed. These measures of reuse were derived by marking up versions of hazard analyses and using a tool for approximate tree matching . These tools were applied to two HAZOP arguments: DUST-EXPERT  and an argument concerned with a mammography system . The latter had been produced by DIRC researchers using a tool that was designed to prompt the analyst with possible arguments already used in similar situations .
In the case of the DUST-EXPERT argument, 256 arguments were classified as "consequence mitigations" of which 203 arguments were unique and 53 involved verbatim reuse. In the mammography analysis 61 arguments were produced of which 56 were unique and 5 used occurrences of verbatim reuse. Using a more relaxed notion of reuse ("trivial reuse"), 56% arguments in the DUST-EXPERT case and 46% of the arguments in the mammography case involved such reuse. While it is not clear what significance can be attached to these statistics insofar as it relates to the confidence which can be attached to an argument, it might lead an analyst to consider these examples of reuse in more detail to ensure that the circumstances are sufficiently similar to provide a similar argument.
1. Clement, T., Cottam, I., Froome, P. and Jone, C. (1999) The development of a commercial "shrink-wrapped application" to safety integrity level 2: The DUST-EXPERT (tm) story. In M. Felici, K. Kanoun & A. Pasquini (eds) 18th International Conference on Computer Safety, Reliability and Security (SAFECOMP 1999), Springer Lecture Notes in Computer Science vol 1698 pp. 216-225.
2. Smith, S.P. and Harrison, M.D. (2002) Improving Hazard Classification through reuse of descriptive arguments. In C. Gacek (ed) ICSR-7 Springer Lecture Notes in Computer Science 2319. pages 255-268.
3. Smith, S.P. & Harrison, M.D. (2005) Measuring reuse in hazard analysis. Reliability Engineering and Safety Science Journal. Vol 89/1 pp 93-104
4. Tsong-Li, J., Zhang, K. Jeong, K. and Shasha, D (1994) A system for approximate tree matching. IEEE Transactions on Knowledge and Data Engineering 6(4): 559-571.
Smith, S.P. and Harrison, M.D. (2002) Improving Hazard Classification through reuse of descriptive arguments. In C. Gacek (ed) ICSR-7 Springer Lecture Notes in Computer Science 2319. pages 255-268.
Smith, S. P. and Harrison, M.D. (2002) Blending Descriptive and Numeric Analysis in Human Reliability Design. Interactive Systems: Design, Specification and Verification. P. Forbrig, B. Urban, J. Vanderdonckt and Q. Limbourg (Eds). Pg 223-237. Lecture Notes in Computer Science 2545. Springer-Verlag.
Smith, S.P. and Harrison, M.D. (2002) Augmenting descriptive scenario analysis for improvements in human reliability design, Applied Computing 2002: Proceedings of the 2002 ACM Symposium on Applied Computing. pg 739-743, ACM:USA.
S.P. Smith and M.D. Harrison, Reuse in hazard analysis: Identification and support, Computer Safety, Reliability and Security (SAFECOMP'03), S. Anderson, M. Felici and B. Littlewood (Eds), Lecture Notes in Computer Science 2788 (2003) 382-395, Springer.
Smith, S.P. & Harrison, M.D. (2005) Measuring reuse in hazard analysis. Reliability Engineering and Safety Science Journal. Vol 89/1 pp 93-104
Michael Harrison (Newcastle) and Shamus Smith
|Page Maintainer: firstname.lastname@example.org||Credits||Project Members only||Last Modified: 10 August, 2005|