AI帮你理解科学

AI 生成解读视频

AI抽取解析论文重点内容自动生成视频


pub
生成解读视频

AI 溯源

AI解析本论文相关学术脉络


Master Reading Tree
生成 溯源树

AI 精读

AI抽取本论文的概要总结


微博一下
The use of Bayesian Networks seems to be promising at different levels of Programmable Electronic Systems dependability analysis, they do not provide a direct mechanism for representing temporal dependencies, that are well implemented in popular techniques for dependability analy...

Improving the analysis of dependable systems by mapping fault trees into Bayesian networks

Reliability Engineering & System Safety, no. 3 (2001): 249-260

引用722|浏览30
EI
下载 PDF 全文
引用
微博一下

摘要

Bayesian Networks (BN) provide a robust probabilistic method of reasoning under uncertainty. They have been successfully applied in a variety of real-world tasks but they have received little attention in the area of dependability. The present paper is aimed at exploring the capabilities of the BN formalism in the analysis of dependable s...更多

代码

数据

0
简介
  • Fault Tree Analysis (FTA) is a very popular and diffused technique for the dependability modeling and evaluation of large, safety-critical systems [1,2], like the Programmable Electronic Systems (PES).
  • The construction of the Fault Tree (FT) proceeds in a top/ down fashion, from the events to their causes, until failures of basic components are reached.
  • In FTA, the analysis is carried out in two steps: a qualitative step in which the logical expression of the TE is derived in terms of prime implicants; a quantitative step in which, on the basis of the probabilities assigned to the failure events of the basic components, the probability of occurrence of the TE is calculated
重点内容
  • Fault Tree Analysis (FTA) is a very popular and diffused technique for the dependability modeling and evaluation of large, safety-critical systems [1,2], like the Programmable Electronic Systems (PES)
  • We first show how an Fault Tree can be converted into an equivalent Bayesian Networks and we show, in Section 4, how assumptions (i)– can be relaxed in the new formalism
  • During Programmable Electronic Systems lifecycle, dependability analysis are performed at different levels addressing hardware, software and the whole system
  • The possibility of use of Bayesian Networks to support the evaluation of overall system dependability along the acceptance process of safety critical Programmable Electronic Systems is under exploration
  • The use of Bayesian Networks seems to be promising at different levels of Programmable Electronic Systems dependability analysis, they do not provide a direct mechanism for representing temporal dependencies, that are well implemented in popular techniques for dependability analysis, such as Markov Chains and Stochastic Petri Nets
结论
  • Conclusions and current research

    Bayesian networks provide a robust probabilistic method of reasoning with uncertainty and are becoming widely used for dependability analysis of safety critical systems as the Programmable Electronic Systems (PES).
  • The authors have dealt with BN versus FT, a very popular technique for hardware dependability analysis.
  • The possibility of use of BN to support the evaluation of overall system dependability along the acceptance process of safety critical PES is under exploration.
  • The use of BN seems to be promising at different levels of PES dependability analysis, they do not provide a direct mechanism for representing temporal dependencies, that are well implemented in popular techniques for dependability analysis, such as Markov Chains and Stochastic Petri Nets
表格
  • Table1: Component failure rate, prior and posterior probabilities
Download tables as Excel
引用论文
  • Henley EJ, Kumamoto H. Reliability engineering and risk assessment. Englewood Cliffs, NJ: Prentice Hall, 1981.
    Google ScholarFindings
  • Leveson NG. Safeware: system safety and computers. Reading, MA: Addison-Wesley, 1995.
    Google ScholarFindings
  • Heckermann D, Wellman M, Mamdani A, editors. Real-world applications of Bayesian networks. Communications of the ACM 1995;38(3).
    Google ScholarLocate open access versionFindings
  • Almond G. An extended example for testing Graphical Belief. Technical Report 6: Statistical Sciences Inc., 1992.
    Google ScholarFindings
  • Portinale L, Bobbio A. Bayesian networks for dependability analysis: an application to digital control reliability. Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence, UAI-99, 1999. p. 551–8.
    Google ScholarLocate open access versionFindings
  • Bobbio A, Portinale L, Minichino M, Ciancamerla E. Comparing fault trees and Bayesian networks for dependability analysis. Proceedings of the 18th International Conference on Computer Safety, Reliability and Security, SAFECOMP99, vol. 1698, 1999. p. 310–22.
    Google ScholarLocate open access versionFindings
  • Fenton N, Littlewood B, Neil M, Strigini L, Sutcliffe A, Wright D. Assessing dependability of safety critical systems using diverse evidence. IEEE Proc Software Engng 1998;145(1):35–9.
    Google ScholarLocate open access versionFindings
  • Torres-Toledano JG, Sucar LE. Bayesian networks for reliability analysis of complex systems, Lecture notes in artificial intelligence, vol. 1484. Berlin: Springer, 1998 (p. 195–206).
    Google ScholarLocate open access versionFindings
  • Malhotra M, Trivedi K. Dependability modeling using Petri nets. IEEE Trans Reliabil 1995;R-44:428–40.
    Google ScholarLocate open access versionFindings
  • Pearl J. Probabilistic reasoning in intelligent systems. Los Altos, CA: Morgan Kaufmann, 1989.
    Google ScholarFindings
  • Neapolitan RE. Probabilistic reasoning in expert systems. New York: Wiley, 1990.
    Google ScholarFindings
  • Dugan JB, Trivedi KS. Coverage modeling for dependability analysis of fault-tolerant systems. IEEE Trans Comput 1989;38:775–87.
    Google ScholarLocate open access versionFindings
  • Garribba S, Guagnini E, Mussio P. Multiple-valued logic trees: meaning and prime implicants. IEEE Trans Reliabil 1985;R34:463 – 72.
    Google ScholarLocate open access versionFindings
  • Doyle SA, Dugan JB, Patterson-Hine A. A combinatorial approach to modeling imperfect coverage. IEEE Trans Reliabil 1995;44:87–94.
    Google ScholarLocate open access versionFindings
  • Wood AP. Multistate block diagrams and fault trees. IEEE Trans Reliabil 1985;R-34:236–40.
    Google ScholarLocate open access versionFindings
  • Knowledge Industries. DXpress 2.0, 1996.
    Google ScholarFindings
  • Microsoft Corporation. Microsoft Belief Network Tools.
    Google ScholarFindings
  • Andersen SK, Olesen KG, Jensen FV. HUGIN — a shell for building Bayesian belief universes for expert systems. Proceedings of the 11th IJCAI, Detroit, MI, 1989. p. 1080–5.
    Google ScholarLocate open access versionFindings
  • Portinale L, Torasso P. A comparative analysis of Horn models and Bayesian networks for diagnosis, Proceedings of the 5th Italian Conference on Artificial Intelligence. Berlin: Springer, 1997.
    Google ScholarLocate open access versionFindings
  • Cooper G. The computation complexity of probabilistic inference using Bayesian belief networks. Artific. Intell 1990;33:393–405.
    Google ScholarLocate open access versionFindings
  • Kjaerulff U. Aspects of efficiency improvements in Bayesian networks. Technical Report. Thesis, Faculty of Technology and Science, Aalborg University, 1993.
    Google ScholarFindings
  • Poole N, Zhang L. Exploiting causal independence in Bayesian network inference. J Artific Intell Res 1996;5:301–28.
    Google ScholarLocate open access versionFindings
  • D’Ambrosio M, Takinawa. Multiplicative factorization of noisy-max. Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence UAI99, Stockholm, 1999. p. 622–30.
    Google ScholarLocate open access versionFindings
  • Delic KA, Mazzanti F, Strigini L. Formalising engineering judgement on software dependability via Belief Networks. Technical Report, SHIP, 1995.
    Google ScholarFindings
0
您的评分 :

暂无评分

标签
评论
avatar
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn