Loughborough University
Browse
05_ARTS_IndepMods_HSJDA.pdf (149.08 kB)

To identify the smallest fault tree sections which contain dependencies.

Download (149.08 kB)
online resource
posted on 2008-09-19, 12:52 authored by H. Sun, J.D. Andrews
Since the early 1960’s fault tree analysis has become the most frequently used technique to quantify the likelihood of a particular system failure mode. One of the underlying assumptions which justifies this approach is that the basic events are independent. However, many systems feature component failure events for which the assumption of independence is not valid. For example, standby dependency, maintenance dependency or sequential dependency can be encountered in engineering systems. In such situations, Markov analysis is required during the quantification process. Since the efficiency of the Markov analysis largely depends on the size of the established Markov model, it is most effective to apply the Markov method only to the smallest possible fault tree sections containing dependencies. The remainder of the system assessment can be performed by the application of the conventional assessment techniques. The key of this approach is to extract from the fault tree the smallest sections which contain dependencies. This paper gives a brief introduction on some main existing dependency types and provides a method aimed at establishing the smallest Markov model for the dependencies contained within the fault tree.

History

School

  • Aeronautical, Automotive, Chemical and Materials Engineering

Department

  • Aeronautical and Automotive Engineering

Citation

SUN, H. and ANDREWS, J.D., 2005. To identify the smallest fault tree sections which contain dependencies. IN: Proceedings of the 16th Advances in Reliability Technology Symposium (ARTS) , Loughborough, UK, 2005, pp. 359-378.

Publisher

© Loughborough University

Publication date

2005

Notes

This is a conference paper.

Language

  • en

Usage metrics

    Loughborough Publications

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC