|
Conclusion
The presence of negated events (i.e., mutually exclusive events) in the FT logic often generates event sequences that are confusing, if not misleading. Disregarding the use of equivalent expressions other than the prime implicants in the FT logic may lead to misinterpretation of the top event contributors. The lack of built-in algorithms in most FT software applications to manage mutually exclusive events compounds the issues of dealing with non-coherent FTs. In addition, a negated event may not fit as the "dual opposite" of the original event. Therefore, an inaccurate quantification will surely follow if non-coherent logic is to be used. The inaccuracy in the top event probability calculation depends on the basic events estimates used, and the FT logic itself. In any case, its importance cannot be predicted until the exact solution is compared.
The BDD approach is promising as a technique for solving non-coherent FTs in an efficient manner. However, the author is not aware of any computer application for the conversion of FT logic into BDDs that has been benchmarked against traditional coherent approaches within the boundaries of a real-world situation.
Unless the system safety practitioner is willing to defray the cost of delving into this cutting-edge topic in terms of time and patience, he or she should refrain from using non-coherent logic for quantitative interpretations in FT analysis, at least for the time being.
About the Author
Sergio (Serg) Oliva has extensive system safety experience including employment as a Senior Safety Engineer at ASCA, Inc. in Rolling Hills Estates, California, and as an analyst in the Probabilistic Risk Assessment group of the Space Shuttle Orbiter at the Johnson Space Center in Houston. He holds an M.S. in safety engineering from Texas A&M University, and is currently pursuing the Certified Safety Professional designation.
References
1. Roberts, N.H., et al. Fault Tree Handbook, NUREG-0492. Nuclear Regulatory Commission, Washington DC, January 1981.
2. Akers, S.B. "Binary Decision Diagrams." IEEE Transactions on Computers, Vol. C-27, No. 6, pp. 509-516, June 1978.
3. Rauzy, A. "New Algorithm for Fault Tree Analysis." Reliability Engineering and System Safety, 40, pp. 203-211, 1993.
4. Andrews, J.D. and S.J. Dunnet. "Improved Accuracy in Event Tree Analysis." In Foresight and Precaution. M.P. Cottam, D.W. Harvey, R.P. Pape and J. Tait, eds. Rotterdam: Balkema, 2000.
5. http://www.cs.virginia.edu/~ftree (Galileo 2.1 Alpha package).
6. http://www.itemuk.com/qras.html (QRAS package).
7. Doyle, S., J.B. Dugan and M. Boyd. "Combinatorial-Models and Coverage:
A Binary Decision Diagram (BDD) Approach." Proceedings of the Annual Reliability and Maintainability Symposium, IEEE, pp. 82-89. Las Vegas: 1995.
8. Tang, Z. and J.B. Dugan. "An Integrated Method for Incorporating Common Cause Failures in System Analysis." NASA Langley Research Center, Contract No. NAS1-02076. Langley, Virginia, 2004.
« previous page
|
