SIGNAL+DRAHT | Ausgabe 12/2012

A framework to improve cause and consequence analysis of hazards in railways

Dezember 2012 | Eric Schöne, Jochen Trinckauf, Michael Kunze, Qamar Mahboob, Ulrich Maschek

Fault Tree (FT) and Event Tree (ET) methods are classical ways of modelling safety risks in railway systems. The FT method is used to model the causes whereas the ET method is used to model the consequences of failures in railway systems. A safety risk model (SRM) is obtained by combining the causes (FT) and the consequences (ET) in the system. This FT and ET based SRM has limitations in modelling systems that are more uncertain and complex. Bayesian Networks (BN) is a probabilistic graphical model, which has several advantages over FT and ET methods. The application of BN to railway safety is novel. In this work, we compare two safety risk models with respect to different modelling and analysis aspects, the information obtained from the analysis and the visualization of the model. It will be investigated whether the modelling and the analysis of safety in a railway system can be improved if BN is utilized.