Digital railway system: <br>collecting big data and making it useful
Technology for data collection that is cost efficient and collects data on regular trains |
One big plus of increasing digitalisation is the ability to collect large amounts of data from which to draw findings. The railway system can take advantage of big data too. The DLR has developed processes and tools for this purpose, from data collection (sensor technology) through fusion and enrichment to data evaluation and utilisation. Central tools for data collection here are inexpensive off-the-shelf solutions which can be used to collect data on regular trains quickly, cheaply and flexibly.
This approach offers an enrichment to the complex and expensive use of test and measurement trains when it comes to inspecting the rail network. At the same time, if data from large numbers of regular train journeys across the entire route network are added to the occasional specific measurement journey, this also means an enormous increase in the quantity of data. If we analyse this data with intelligent algorithms, we could see a quantum leap in the maintenance and repair of the rail infrastructure through the use of digitalisation if maintenance can be undertaken as needed instead of at fixed intervals. This reduces costs and makes the planning of personnel and replacement parts more efficient, preventing loss of service and thus delays.
The key to this big data approach: the technology required for data capture fits into a small suitcase. This includes a software platform which records the data, evaluates them in real-time and communicates filtered information to a central data platform. A huge range of sensors such as cameras, microphone, inertial sensor technology or GNSS based location components can be connected as needed. This enables very frequent collection of data. For example, vibrations can be measured and used to spot faults in the rails. Or, with an integrated microphone, the level of noise can be recorded which can provide a network-wide overview of noise emissions if large amounts of data are collected. Viewed over time, this too can enable conclusions to be drawn about damaged rails. Measurements concerning energy use along the track can highlight potential areas of optimisation for driver assistance systems. Constant provision of vehicle location data permits a real-time view of the total load on the network and the direction of travel of the trains. This will also lay a foundation stone for the upcoming automation of the rail system. The data recorded are transmitted to a server via GSM permitting data analysis of all data over a longer period of time. Thanks to long-term data collection we are within reach of a future prognosis of infrastructure changes. The software comes packed in a case, making it very versatile to use but it can also send current position data from a smartphone. Or, integrated into two engineering works lights, it can be used to mark the beginning and end of engineering works which can then be sent in visual form to rail service managers and train drivers as hazard points.
