The introduction of predictive maintenance management is currently a key component of many digitalisation strategies for railway infrastructure operators worldwide. Faults in the track superstructures, rails and control and safety technology incur great maintenance costs year upon year, accounting for around 50 percent of the total life cycle costs of railway infrastructure in Europe. Predictive maintenance management provides the only means of significantly reducing costs. The basic prerequisites for this are continuous, automatic condition monitoring of the relevant assets during operation, as well as the ability to automatically diagnose and predict the status of the system based on an extensive and complex pool of data (Big Data). The German Aerospace Center (DLR) is working with the rail industry to find ways of tackling these challenges in national and international research projects, as well as in the EU joint undertaking Shift2Rail. DLR’s research activities extend across the entire process chain – from embedded sensors through to visualised status information. This includes multi-sensor systems on rail vehicles and in signal boxes, data management and data fusion, as well as the development of appropriate algorithms based on state-of-the-art data science and artificial intelligence methods. At InnoTrans, DLR will be showcasing its progress in embedded rail monitoring with conventional rail vehicles, switches and cable systems for electronic interlocking stations.