Application of importance measures to transport industry: Computation using Bayesian networks and Fault Tree Analysis

Author

Mahboob, Qamar ; Schöne, Eric ; Kunze, Michael ; Trinckauf, Jochen ; Maschek, Ulrich

Author_Institution

Dept. of TU Dresden, Railway Signaling & Transp. Safety Technol, Dresden, Germany

fYear

2012

fDate

15-18 June 2012

Firstpage

17

Lastpage

22

Abstract

Importance measures are useful in the identification of components, which are most effective towards safety improvement. This paper will summarize a number of Component Importance Measures (CIM), computation of the CIM using Fault Tree Analysis (FTA) and Bayesian Networks (BNs) will be investigated for the transport industry. The BNs are directed acyclic probabilistic graphical models, used for joint distribution of random variables in a concise and efficient way. The BNs have several advantages over classical ways like FTA.

Keywords

belief networks; fault trees; probability; railway industry; railway safety; transportation; BN; Bayesian networks; CIM; FTA; component identification; component importance measures; directed acyclic probabilistic graphical models; fault tree analysis; railway industry; random variable distribution; safety improvement; transport industry; Bayesian methods; Computational modeling; Computer integrated manufacturing; Rail transportation; Random variables; Reliability; Safety; Bayesian Networks analysis; Fault Tree analysis; importance measures; transport industry;

fLanguage

English

Publisher

ieee

Conference_Titel

Quality, Reliability, Risk, Maintenance, and Safety Engineering (ICQR2MSE), 2012 International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4673-0786-4

Type

conf

DOI

10.1109/ICQR2MSE.2012.6246180

Filename

6246180