Reliability Analysis of Dynamic Systems by Translating Temporal Fault Trees into Bayesian Networks
Classical combinatorial fault trees can be used to assess combinations of failures but are unable to capture sequences of faults, which are important in complex dynamic systems. A number of proposed techniques extend fault tree analysis for dynamic systems. One of such technique, Pandora, introduces temporal gates to capture the sequencing of events and allows qualitative analysis of temporal fault trees. Pandora can be easily integrated in model-based design and analysis techniques. It is, therefore, useful to explore the possible avenues for quantitative analysis of Pandora temporal fault trees, and we identify Bayesian Networks as a possible framework for such analysis. We describe how Pandora fault trees can be translated to Bayesian Networks for dynamic dependability analysis and demonstrate the process on a simplified fuel system model. The conversion facilitates predictive reliability analysis of Pandora fault trees, but also opens the way for post-hoc diagnostic analysis of failures.
Unable to display preview. Download preview PDF.
- 2.Boudali, H., Dugan, J.: A new bayesian network approach to solve dynamic fault trees. In: Proceedings of Annual Reliability and Maintainability Symposium, pp. 451–456 (January 2005)Google Scholar
- 3.Bruns, G., Anderson, S.: Validating Safety Models with Fault Trees. In: Górski, J. (ed.) SAFECOMP 1993, pp. 21–30. Springer, London (1993)Google Scholar
- 4.Cozman, F.: JavaBayes (2001), http://www.cs.cmu.edu/~javabayes/
- 5.Dugan, J.B., Bavuso, S.J., Boyd, M.A.: Fault Trees and Sequence Dependencies. In: Proceedings of Annual Reliability and Maintainability Symposium, pp. 286–293 (January 1990)Google Scholar
- 10.Marquez, D., Neil, M., Fenton, N.: Solving Dynamic Fault Trees using a New Hybrid Bayesian Network Inference Algorithm. In: 16th Mediterranean Conference on Control and Automation, pp. 609–614. IEEE (2008)Google Scholar
- 14.Montani, S., Portinale, L., Bobbio, A., Varesio, M., Codetta-Raiteri, D.: A tool for automatically translating Dynamic Fault Trees into Dynamic Bayesian Networks. In: Annual Reliability and Maintainability Symposium (RAMS 2006), pp. 434–441. IEEE (2006)Google Scholar
- 15.Pearl, J.: Probabilistic reasoning in intelligent systems: Networks of Plausible Inference. Morgan Kaufmann (1988)Google Scholar
- 16.Vesely, W., Dugan, J., Fragola, J., Minarick, R.J.: Fault Tree Handbook with Aerospace Applications. Tech. rep., NASA office of safety and mission assurance, Washington, DC (2002)Google Scholar
- 18.Walker, M.D.: Pandora: A Logic for the Qualitative Analysis of Temporal Fault Trees. Ph.D. thesis, University of Hull (2009)Google Scholar