Back1 Back2 Back3 QSAV

Quality assurance for the German toll system


Quality assurance for the German toll system


Project key data

6 people

30+ months

Big data QA platform

Functional Programming

Scala, Python

Cassandra, Kafka, PostgreSQL

Why we like showing this case

You see, you see nothing: the German truck toll has been running safely and smoothly for over 18 years - even though the quality assurance of a big data application that collects tolls for more than 4 million vehicles every day is a highly complex task. We have contributed quite a bit to this. Making complex systems manageable is the most exciting challenge for us!

After all, the toll system makes a vital contribution to infrastructure financing in Germany.

Task and solution

The German toll system not only collects tolls for more than 4 million vehicles, but also has to record different types of vehicles, from 3.5-ton trucks to heavyweights of more than 12 tons. This generates many terabytes of data, which are processed completely anonymously in a smart chain. The technology must be scalable and capable of adapting to constant technical and operational change. Quality control must be carried out to ensure that the system works at all.

Applying best DevOps practices, the Ambrosys team is constantly working with state-of-the-art methods and tools to ensure the operation of the QA platform. Statistical measures are applied to mass data in order to be able to reliably assess the entire processing chain from pre-processing to the final result. Among other things, they allow data integrity to be checked, data errors to be identified and corrected and deviations and outliers to be analyzed. These steps are crucial to ensure that the calculated tolls are correct and that the system works smoothly every day. Eventually, monitoring indicates the quality of the results and allows the system owner – our customer – to initiate technical and business measures to ensure or improve quality.

Our system includes business analysis capabilities and even allows to explore other business areas.

Smart tolling – enabler of sustainable mobility!

"Toll" has a bad rap. We would like to set the record straight!

In order to align mobility with the needs of people and our planet in the future, we need intelligent technologies. Traffic space is a scarce resource, especially in cities. If we put a price on this resource, we can ensure fair distribution. An intelligent, socially balanced toll may even be the only way to maintain individual mobility and avoid strict bans. Ambrosys uses intelligent technologies to create the conditions for policy-makers to shape our mobility in a way that is socially just, economically viable and environmentally friendly. Intelligent tolling is route-based, transparent, easy to use, and can be dynamically adapted to the situation using advanced algorithms.

In the near future, more and more cities will take the step towards a city toll. Ambrosys advises them holistically on the possible options and provides both mobility expertise and a network of partner companies to develop complete solutions tailored to local needs.

The importance of good devops practices

Effective DevOps practices are essential for making complex software systems reliable, adaptable and efficient. For technology companies, this means that product innovations can be on the market faster and costs can be reduced. Unfortunately, there is no short, exhaustive explanation of the term. Good DevOps practice results from a range of methods that interlock and fundamentally improve collaboration between IT development and IT operations, such as:

Continuous Integration and Continuous Deployment ensure that new features reach the production environment quickly but in a controlled manner and that integration issues are detected early.

Monitoring ensures rapid detection of deviations and glitches during operation.

Versioning systems make changes transparent and traceable and enable parallel working.

DevOps are not individual components, but a "best practice" set of behaviors, and as the old saying goes: best practice is an output, not an input. They sure require system components, but they must be trained and applied by everyone involved.

We set up the processes and components required for DevOps for our customers - preferably as part of a development team, but if necessary also as coaches, teaching how to apply them. This way, our customers benefit from the many years of experience we have gained with DevOps in various organizations and build up a high level of skills in their own organization.


→ Questions or thoughts about smart mobility, intelligent tolling or DevOps? Dr. Markus Abel is looking forward to your message.

Go back