Safety is an essential requirement for railway transportation. There are many methods that have been developed to predict, prevent and mitigate accidents in this context. All of these methods have their own purpose and limitations. This paper presents a new useful analysis technique: timed fault tree analysis. This method extends traditional fault tree analysis with temporal events and fault characteristics. Timed Fault Trees (TFTs) can determine which faults need to be eliminated urgentlyand it can also provide how much time have been left at least to eliminate the root failure to prevent accidents. They can also be used to determine the time taken for railway maintenance requirements, and thereby improve maintenance e ciency, and reduce risks. In this paper, we present the features and functionality of a railway transportation system, and principles and rules of TFTs. We demonstrate the applicability of our framework by a case study on a simple railway transportation system.