by Kai Höfig, Dominik Domis, Mario Trapp, Heiko Stallbaum
Abstract:
The growing complexity of safety-critical systems is leading to an increased complexity of safety analysismodels. Automatic transformations from the system model into the safety analysis model help to reduce timeand cost for safety analyses. In this paper, we describe a pattern annotation framework for development modelsof safety-critical systems that is used to annotate system entities of fault tolerance mechanisms. We use theannotations as a source of semantic information to automatically generate parts of the safety analysis model. Thepresented approach contributes to the safety analysis domain by explicitly preserving the knowledge of domainexperts and engineers within the model. Our approach allows performing a semi-automated preliminary safetyanalysis of fault tolerance mechanisms to estimate the failure behavior of a safety-critical systems architecturemodel in early stages of the development. This saves effort and costs for ongoing steps of the developmentprocess and prevents setbacks in the development.
Reference:
K. Höfig et al., "Pattern-based Safety Engineering: Semantic Enrichment of System-Architecture Models for Semi-Automated Safety Analysis", in ESREL, Proceedings of, 2010.
Bibtex Entry:
@INPROCEEDINGS{Hoefig2010,
author = {Höfig, Kai and Domis, Dominik and Trapp, Mario and Stallbaum, Heiko},
title = {Pattern-based Safety Engineering: Semantic Enrichment of {System-Architecture}
Models for {Semi-Automated} Safety Analysis},
booktitle = {ESREL, Proceedings of},
year = {2010},
abstract = {The growing complexity of safety-critical systems is leading to an
increased complexity of safety analysismodels. Automatic transformations
from the system model into the safety analysis model help to reduce
timeand cost for safety analyses. In this paper, we describe a pattern
annotation framework for development modelsof safety-critical systems
that is used to annotate system entities of fault tolerance mechanisms.
We use theannotations as a source of semantic information to automatically
generate parts of the safety analysis model. Thepresented approach
contributes to the safety analysis domain by explicitly preserving
the knowledge of domainexperts and engineers within the model. Our
approach allows performing a semi-automated preliminary safetyanalysis
of fault tolerance mechanisms to estimate the failure behavior of
a safety-critical systems architecturemodel in early stages of the
development. This saves effort and costs for ongoing steps of the
developmentprocess and prevents setbacks in the development.}
}