Abstract: ′Fail-operational′ is the highest requirement for urban rail transit signaling system, and also an understanding of system safety from the high-level operation. With the trend of automatic systems expanding their function boundaries, human intervention is no longer considered a safe and reliable means, and the system itself should be capable of treating more anomalies. Concepts such as fault tolerance, safety integrity level and resilience related to system reliability are analyzed. System resilience theory is introduced into the design of urban rail transit signaling system, and system resilience is quantitatively calculated using system ability loss. A design model of the resilience system is established, and is then applied to the design of intelligent scheduling system with resilience measurement as the optimal control target. Simulation verification proves that the system resilience theory and method are effective in urban rail transit signaling system design.