Train control uncertainty treatment and safety assurance shifting to automation


Author: Fei Yan
Co-authors: Chunhai Gao, Ru Niu, Tao Tang
Day: Aspect Day One
Session: Safety

Large cities depend heavily on their metro systems to alleviate traffic jam, while disruptive incidents have become more common, causing threats to passenger safety and transport service plan. Adopting a fully automatic operation systemFAO with advanced technology, stable performance, efficiency first has become the urgent need of global rail transit construction. The objective of this paper is to show how to deal with train control uncertainty in driverless metro and give a possible solution to realize a resilience metro system. The paper selects a FAO system scenario to use causal scenario search algorithm which based on Systems-Theoretic Process Analysis (STPA) for safety analysis, obtaining more causal scenarios and related safety requirements. At the same time, it discusses the safe process of testing, verification and validation and the interaction between passengers and the system. The safety assurance scheme of the FAO system in the design and development process is proposed.Index Terms—Fully Automatic Operation System, System Assurance, Uncertainty Treatment