To ensure that all Deutsche Bahn trains are ready for service at any time, they are inspected on a regular basis. Digitalisation is also making inroads into this maintenance process: by means of vehicle intelligence in conjunction with a data integration and analysis platform, the reliability and availability of the locomotives are being optimised. This is expected to reduce costs, for example, by cutting the number of visits to the workshop.
For this purpose, DB Cargo has been collaborating with DB Systel since 2015 to develop a successful Operational Intelligence solution for the TechLOK project that will permit comprehensive use of the vehicle data. In the course of this work, freight locomotives are being fitted with sensors and diagnostic boxes in order to record the relevant vehicle data. This data is sent via the mobile phone network to a special receiver component in a central computer, where it is automatically converted and indexed.
Preparation and use of the data
Further analysis is then performed on this data integration platform, enabling conditions, such as an excessively high coolant temperature, to be analysed. Using such simple logic for the initial locomotive classes, alarms can also be triggered in the event of critical error codes and linked to recommended courses of action. In addition, more complex visualisations will be possible within the near future, for example, to compare the energy consumption of different locomotives at a glance.
The challenge for these tasks was to link the different interfaces and needs to one another. Because, for a long time, there was no central dashboard as an overview for recording and visualising different system data. For this purpose, DB Systel implements the Splunk system and collaborates with the customer to optimise it for the specific use.
This enables different machine data to be integrated into just one platform. In future, it is intended that this machine data will also be linked with other data, such as weather information.
Broad range of use
For other customers as well, the use of this operational intelligence brings nothing but advantages. For example, they are provided with central cockpits for scheduling and sales analysis, for monitoring systems or overviews of faults. In addition, thanks to a simple search language and a large app community, the system can be quickly integrated into different workflows.
Splunk has also been in pilot operation since the end of 2015 at DB Sales, where it is connected to the NVS sales system so that travel centres and partner companies can now view and evaluate their data independently and on a daily basis.
Advantages in all areas
DB Cargo is now reaping the benefits of the first positive experiences gained from the use of Splunk. Ultimately this will significantly simplify vehicle monitoring, because the different sensor and diagnostic data can be checked in real time. In the current configuration, the system already receives vehicle data from around 500 locomotives spread across five classes.
The advantage is clear: dispatchers, fleet managers, workshop staff and engineers can access and analyse the machine data of the respective locomotive at any time – whether it be for the current point in time or for a period in the past. All of this data is of great importance for operations and for operational business processes – and in the long term, too. The information obtained is stored for business intelligence, i.e. historical data analysis and reporting, or for a reliable prediction, and corresponding tools can be made available.
Four levels of maintenance
Intelligent use of sensor technology and the associated analysis tools now offer many possibilities for optimisation. This is illustrated by the example of maintenance:
Vehicles today are generally maintained on a reactive maintenance or preventive maintenance basis. This means that a maintenance task is only carried out in the case of an actual fault or according to fixed intervals. These approaches, however, offer only limited opportunities for optimisation.
In future, digitalisation should offer vehicles the option, wherever practicable, of condition-based maintenance or predictive maintenance. This means that maintenance work is only carried out if the condition of a component demands such action, or if the systems detect the impending or even the actual failure of a component. This approach maximises the availability of rolling stock.
Integration of supplementary services
The sensor data collected can be further processed and enhanced with supplementary data in order to obtain meaningful results. Additional data, such as temperature and humidity, is made available, as well as that relating to mechanical incidents. Integrated map services, for example, containing information about the infrastructure of the Deutsche Bahn rail network, can also be used in the analyses of the machine data, enabling the customer to pinpoint the location of fleet locomotives at any time. But the DB Systel system offers many additional options for supporting the business processes in the overall context – for example, by forecasting the service life, facilitating the quality assurance of vehicles, or continuously analysing possible vulnerabilities.
The expansion is progressing
So soon after starting, the DB Cargo reference project TechLOK and the use of innovative technologies is already proving extremely successful, and its expansion is proceeding rapidly. By 2019, about 2000 vehicles from more than 16 classes will be included in the scheme. Work is already in progress on further possible applications, for example, the integration into Splunk of freight wagons at DB Cargo.
But this is by no means the limit, because the process is fully transferable: to train doors at DB Regio or to IC trains at DB Long Distance – wherever sensors record machine data, this platform can visualise this data in real time, thereby supporting important decisions in the operational process. Because that is what DB Systel is all about: working together to achieve the best results.