Reliability assessment of CTCS-3 using Bayesian networks

Train control system of high speed railway is the core equipment and technology to ensure the safe and reliable operation of high-speed trains, so reliability assessment of train control system counts for much. In this paper, firstly, Bayesian Networks (BN) technology is introduced for the limitation of traditional fault tree analysis (FTA) method. Secondly, reliability model of CTCS-3 train control system is established by BN transformed from the fault tree (FT) of CTCS-3. Finally, reliability of CTCS-3 and its redundant configuration is assessed by BN´s advantage of expressing uncertainty knowledge and its bidirectional reasoning ability. It is not only convenient to identify the failure combination mode leading to the failure of the train control system, but also be able to find out the weak links in CTCS-3 system. Through the case analysis, BN shows great superiority in reliability assessment of system compared with FTA. What is more, it provides a new foundation for reliability assessment of train control system.