Towards modeling and evaluation of availability of communication based train control (CBTC) system

This In this paper a stochastic automation networks (SANs) model for the computing the availability of communication based train control (CBTC) system is developed, which captures the lossy radio communication link and the transmission procedures based on 802.11 protocols. A scaled generalized minimum residual (SGMRES) iterative method is provided to reduce the ill-condition of steady-state equation and the state-space explosion problem. The availability (mean time between train stop, MTBTS) of CBTC is quantitatively evaluated and compared under different deadlines and time between two successive train position reports. Analytic results are conducted to verify that the required mean time between train stop (MTBTS) of 250,000 hours is met at the deadline of 2 minutes.