Which data does DB Regio provide?

• Vehicle location data (see example files)
• Time table information (RIS-API)
• Track data
• Further to be discussed during Proof of Concept (PoC)

Is RINEX data available?

• Unfortunately not.

Which data do you receive on your trains eg. GPS L1/l2, GLONASS L1/L2…?

• We only receive the processed GPS-data results in WGS 84.

Is it possible to get GNSS raw data?

• Unfortunately, we can not access this data in our backoffice.

Can we increase GPS quality by adding our own hardware or upgrading the current technology in the vehicles?

• No, thats exactly the challenge.

Whats the tricky part about match mapping?

• The parallel tracks in our depots are very close to each other.
• It is challenging to increase the accuracy to match the train to the right track.
• Check the file  “Example of depot tracks”.
• Further information: It is not possible to update the software of our existing systems.

Where is the graffiti removed from the trains?

• Depending on the size and planned delivery to the factory, cleaning takes place in the workshop or on storage tracks.
• Our focus is on mobile graffiti removal, in other words, without a fixed power or water supply.
• Your solution must also be applicable with a maximum of two workers required to remove the graffiti.

What cleaning agent is used?

• We must use certain certified cleaning agents.
• These are applied to the vehicle, absorbed for a few minutes, and then removed.
• The cleaning agent and the paint combine to form a jelly-like mass.

What is the purpose of the risk score?

• Due to unforeseen situations prior to train deployment, journeys may not always start as scheduled.
• Deviations are often detected too late and sometimes not at all.
• With the risk score, we want to identify the likelihood of a schedule deviation and use it to actively alert our dispatchers.

What data is provided?

• Possible data includes GPS, schedules, operational and traffic trips, and workshop planning.
• What other existing data sources might be of interest to you to better assess the risk, is part of the PoC.

What is an annual shift schedule?

• Annual shift schedules are long-term deployment plans that specify when and to what extent employees should work in shifts over an entire year.

What is the content of the PoC?

• During the onboarding week starting June 16, we define the specific scope of the 100 days.

The content could include, for example:

• Create a forecast of personnel availability for 2026, including all known absences and attendances (rest levels, annual shift schedule, vacation, part-time work due to retirement, training needs, etc.).
• Create a proposal to automatically adjust days off and the annual shift schedule individually or as a whole (blocking days with low availability for time off, training, etc.).
• Evaluate shift patterns at deployment sites and, if known, take construction work into account.

Which tanks should be equipped with sensors?

• The requirements vary depending on the model series.
• Basically, we are looking for a solution that can measure the fill level of diesel and AdBlue tanks, as well as cooling, freshwater and wastewater tanks, and transmit this to our control center.

Who receives the data?

• Data should be able to reach maintenance and dispatch personnel.

Which model series should be equipped first?

• Although this is not yet confirmed, it will likely be the Regio 623, 642, or 643 series.

Is it permissible to drill into the tank?

• The fewer changes to the train, the better. Changes require approval. If possible, we should avoid drilling through existing walls and tanks.

Which series should be equipped first?

• Although not yet confirmed, the Bombardier 442 series is the most likely candidate.

Who receives the data?

• Your team should be able to identify factors which will lead to the failure of battery modules and process the data accordingly to provide a prognosis to maintenance teams.

What should be measured and processed?

• All values ​​that could indicate a critical battery condition (such as internal resistance, temperature, voltage and charge history).

Do you have technical information on the installed batteries?

• Lead-acid batteries, 12V blocks, 80 - 105 ampere-hours. Connected in series (9*12V = 108V system).

We identified 3 use cases that focus on process automation, please tell us in you application which use cases you apply for.

Use Case 1 – Wohin Du Willst

• For our platform "wohin-du-willst.de" we receive tariff information from our partners in various formats (Access,Excel, Init, PDF) mostly via email.
• This information should be automatically extracted from these files and transferred to templates (Excel, CSV), so that manual input is no longer necessary.
• Automated plausibility checks must be carried out and the data must to be verified with a simulation of a test journey in the platform.

Use Case 2 – Review of incident coding

• There are codes for disruptions during operation.
• You should analyse all codes from the source system and check their plausibility by comparing data with other systems.
• Your intelligent solution also understands the context in text boxes and converts this into a predefined incident code.
• Last but not least, you take over the manual step of submitting a change request in the source system, fully in line with process automation.

Use Case 3 – Converter und content delivery system documentation manuals

• The aim of this use case is to automate the process flow from technical documentation to checklists, to enable the further development of content and to prepare the content in a user-specific way (legal requirements vs. practical needs).
• Intelligent data processing enables the system in question to use a wide variety of input formats and structures to create standardized, target group-specific output formats such as checklists.
• Adaptation requirements for source data (e.g. from manufacturer) are recorded and transmitted directly by the end user.

Do you have any details about input formats?

• The manuals are created in the manufacturer's editing tool.
• Mainly XML files are used, based on a DTD structure of the manufacturer.
• The aim is to define a suitable structure for data exchange and to ensure that this enables standardised exchange between the manufacturer and DB system via an API.

Which data needs to be analysed?

• We are looking at operating and maintenance manuals (relevant and technical information documented by the manufacturer for the use and maintenance of their products).
• They contain technical information on the use and maintenance of products, e.g. on systems, components or installations.
• They also take into account legal requirements, standards and process-specific requirements.
• The content consists of various forms of data.

Do you have an example of a structured output?

• We uploaded a file which equals our standard work constructions

What format should the output be in? Excel or structured formats such as XML or JSON?

• The main focus is on structured formats such as XML (according to XSD) (or JSON, for example, so that the data can be fed directly into a third-party system).
• Excel or PDF are then created by the editorial system and the target system of the DB itself.