Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Efficient reachability graph development for qualitive analysis of state/event fault trees

: Roth, Michael; Hartoyo, Agus; Liggesmeyer, Peter


Institute of Electrical and Electronics Engineers -IEEE-; IEEE Computer Society:
IEEE International Symposium on Software Reliability Engineering Workshops, ISSREW 2015. Proceedings : 2-5 November 2015, Gaithersburg, MD
Piscataway, NJ: IEEE, 2015
ISBN: 978-1-5090-1943-4 (Print)
ISBN: 978-1-5090-1944-1
International Symposium on Software Reliability Engineering (ISSRE) <26, 2015, Gaithersburg/Md.>
Fraunhofer IESE ()
software; petri nets; software engineering; qualitative analysis; Fault Tree Analysis (FTA); Quantitative Analysis; State Event Fault Tree (SEFT); fault tree; reliability analysis

In the modern environment, people are more and more dependent from software-controlled systems. If it comes, however, to the certification of those systems, frequently used methods like fault tree analysis or Markov models have some crucial disadvantages, especially in modeling the hierarchical decomposition as well as the temporal behavior of the underlying system. To deal with these weak points, state/event fault trees (SEFTs) [9] were developed. All analysis methods of SEFTs - qualitative and quantitative - based on their reachability graphs which are traditionally developed by translation into equivalent Petri-nets. However, this Petri-net translation and its subsequent reachability graph development is a computationally expensive procedure that comes mostly with a state-space explosion. In this paper, we propose an approach that is able to develop an SEFT's reachability graph directly without an intermediate Petri-net model. In the evaluation we have carrie out some case studies in order to show that the direct approach increases the efficiency w. r. t. memory and time consumption while maintaining the same results as the traditional approach.