Regional Traffic 2026

It is part of the 100 days to analyze various data sets and data sources to identify which mix of data causes best results.

The following data, among others, is available for analysis: vehicle malfunctions from SAP, deviation information from the timetable, malfunction codes, malfunction inventories for individual vehicles, and social media.

Incidents are documented with so-called fine codes, which classify the causes of the incidents. Based on your data analysis, the appropriate DB Regio fine code should be assigned automatically.

For the PoC, we are looking at the BR 423. However, it is desirable that the solution can be used for as many vehicle types as possible.

We are looking for easy-to-install sensors that detect abnormalities in door behavior and predict how quickly the malfunction will worsen. As a productive system, the information should be sent to the control center system and to SAP as a maintenance message. In the first step, a dashboard is sufficient.

Train personnel can digitally record vehicle damage in DB system with a unique damage code. The quality of the findings is not sufficient for legally binding documentation or damage prognosis. Also, there is a lack of capacity to create damage reports, and information on vehicle location is not available. In order to improve the quality of the data transmitted to the authorities, a high level of manual post-processing is currently required.

Bodycams worn by security personnel in parking lots record initial graffiti findings on vehicle exteriors, including vehicle number (UIC no.), parking location, and time, and transmits these results as one event/data packet via real-time communication tool between Bodycam and IDA (Intelligent digital assistant- DB tool). In IDA, the events/data packets are mapped with the findings from circulation and integrated into the existing data flow to the Corporate Security Cloud.

We are using Bodycams from netco. https://www.netco.de/body-cam/

We consider the planning of movements and parking. This refers to vehicle movements after the last passenger journey and before the first passenger journey. During this time, a vehicle passes through several stations, including water disposal, exterior/interior cleaning, refueling, maintenance, and parking.

We are looking for a georeferenced digital map with traffic attributes that is particularly suitable for bus transport and forms the basis for further applications. It is important to have high-performance real-time display and compression of very large quantities of moving objects, which must also be reflected in the UI/UX. 

This map must be editable so that road closures and diversions can be incorporated manually. Buses are displayed as digital twins in real time in the system. The updated route is then transmitted to the bus driver in real time. 

In addition, the system should be intelligent enough to use historical data to make schedule-based predictions about the location of the bus, even if it is not currently sending its location. 

It should be possible to derive context-related dispatching measures, such as suggesting alternatives directly based on the route, but also to set the routes yourself. The system also recognizes correlations with connecting trips and informs the dispatcher that a connecting connection cannot be maintained as planned.

Extended bus route data (stops), dynamic content (road closures, stop closures), configuration (bus navigation), bus status (e.g., passenger counts, disruptions), but especially position data, timetable data, forecasts, and driving data specifications.