Internet Explorer, Chrome Browser, Firefox Browser, Safari Browser
Feature21 April 2017,
updated09 October 2017Siemens Mobility GmbHMunich
Utmost reliability and maximum availability are critically important for ensuring the cost-efficient operation of rail vehicles and the infrastructure they use. After all, malfunctions and downtimes cost money, cause delays and frequently also lead to compensation claims from passengers, local transport purchasers and freight customers. Long before faults actually occur, their potential sources should be identified. To provide this information, Siemens is the first company in the rail industry to operate a special data analytics center, located in Munich, Germany.
Press Pictures
Siemens networks locomotives of Deutsche Bahn
DB Cargo AG has commissioned Siemens to equip its locomotive fleet for condition-based, predictive maintenance. The upgrade applies to Siemens locomotives of the 152 Series Eurosprinter ES64F and locomotives of the 170 and 191 Series, both Vectron types. For the 152 Series locomotives, Siemens will retrofit the necessary telemetric systems and network all locomotives with the "TechLOK" system used by DB Cargo. The contract has a term of six years. The telemetric systems continually collect data on the condition of the locomotives. With this data, experts at Siemens' Mobility Data Services Center will work with DB Cargo to develop identified applications and data analytics models.
In the picture: Vectron at DB Cargo Polska
In the picture: Vectron at DB Cargo Polska
Infographics
Heading for Data-Driven Rail Systems
Long before faults actually occur, their potential sources should be identified. To provide this information, Siemens is the first company in the rail industry to operate a special data analytics center, located in Munich, Germany. In this Mobility Data Services Center, a team of data scientists, physicists, engineers, computer scientists and mathematicians thoroughly analyzes the diagnostic data being collected from rail vehicles and rail line infrastructure components. Algorithms and models are worked out using machine learning, data analytics, mathematical and physical methodologies to provide secure forecasts for the future behavior of vehicles and components. In this case, “secure” means a probability of well over 90 percent that the forecast will be accurate. Inaccurate forecasts also cost money and unnecessary downtimes.