The ANYMOS project

To make public transport more sustainable and efficient, decision-makers need accurate insights into how many passengers travel on which routes, at what times and with which types of tickets. Recording passenger flows and creating so-called origin-destination-matrices therefore form an important basis for planning and optimising operations and services.

Although solutions now exist that generate such data directly through use – such as checkin/check-out (CICO) and be-in/be-out (BIBO) systems – not every public transport operator has access to these technologies. Empirical passenger surveys, in combination with automatic passenger counting (APC) systems, remain a common way of determining and analysing passenger flows. However, passenger surveys have their drawbacks, including sampling distortions, limited timeliness, poor repeatability and high time and financial efforts due to manual implementation. Alternative methods for reconstructing passenger flows are therefore crucial to close these gaps.

Anonymous and data protectioncompliant processing

As part of the recently completed ANYMOS research project, methods were developed to reconstruct passenger flows based on connection requests from a passenger app, ticket sales data, passenger count results and other information such as demographic data (census data from the German Federal Statistical Office). Together with its project partners – transport association Karlsruher Verkehrsverbund (KVV) and the Research Centre for Information Technology (FZI) – INIT processed and evaluated this information. A key focus of the ANYMOS project was the anonymised and data protection compliant use and further processing of mobility data.

From booking and connection enquiries to passenger flows

The basis for determining passenger flows was KVV’s multimodal mobility platform regiomove, for which INIT had provided the booking and payment platform. Entries made by the around 64,000 users in the regiomove app’s information interface were recorded and stored. These entries, or log files, contain details about the start and destination of the request, the time of the query and the geo-coordinates of the location from which it was made.

Next, these data were merged with current ticket purchases from INIT’s ticketing back-office system. To comply with data protection regulations and prevent the re-identification of personal information, anonymisation methods were developed in collaboration with FZI – without compromising the accuracy of passenger flow analysis.

INIT also developed the system architecture for ANYMOS. This enables the automated retrieval of user data from the regiomove app (KVV), the anonymisation of data (FZI) and its processing (INIT). The anonymised data were then stored in a database and – as far as data protection regulations allow – combined with other datasets and visualised in a front-end dashboard for KVV.

Cross-referencing with other data, such as demographic data (age, place of residence, household size), enabled the researchers to verify statistical representativeness. This makes it possible to draw appropriate conclusions about passenger flows.

Outlook: Dynamic visualisation of passenger flows

Visualization of passenger flows in the form of green lines on a geographic map

Several dynamic visualisations were developed in ANYMOS to illustrate passenger flows. Passenger movements can be aggregated and displayed at various spatial levels – from city districts down to individual stops. The thickness
of the arrows represents the number of connection requests from one district to another (see image).

Origin-destination-matrices and bar charts offer an excellent overview, showing comparisons between connection requests and tickets sold. The data can be aggregated daily to track how demand evolves over time. This enables all key data on requests and ticket sales to be viewed quickly and clearly at any time.

Conclusion

The findings of the research project will help tailor KVV’s services even more closely to actual demand in future, further strengthening public transport as an attractive, flexible and sustainable means of travel.

Contact

Dr. Jochen Wendel

Senior R&D Manager,
Deputy Team Manager

init SE

Germany

Contact

Subscribe now

This article was published in our customer magazine INITiative, issue 3/2025. If you want to receive the magazine regularly per e-mail, we kindly ask you to subscribe.

Proceed

Service planning based on travel patterns