ANYMOS: Smart service planning in Public Transport using anonymized mobility data
Every day, thousands of passengers move through the network - but who travels when, where, and with which type of ticket?To plan public transport efficiently, operators need accurate insights into passenger flows.
In the ANYMOS research project, INIT worked together with the Karlsruhe Transport Association (KVV) and the FZI Research Center for Information Technology to develop innovative methods for extracting this information from existing mobility data - fully compliant with data protection regulations - even in areas without traditional check-in/check-out or be-in/be-out systems.
At the heart of the project is a flexible system architecture that automatically integrates, anonymizes, and analyzes multiple data sources. These include journey requests from the multimodal platform regiomove, ticket sales data, and results from automated passenger counts. Additional demographic data help to statistically classify and validate the results.
A key focus of ANYMOS is protecting personal data. In close collaboration with FZI, procedures were developed that allow meaningful demand analyses without revealing information about individual passengers. The processed data can be visualized clearly: passenger flows are made transparent across different spatial levels - from neighborhoods to individual stops. Origin-destination matrices and time comparisons help transport operators identify demand trends faster and adjust their services more precisely.
Another exciting aspect of the project focused on AI-based pseudonymization of in-vehicle video surveillance. Sensitive areas of the footage are protected in real time, while anonymized analyses remain possible for planning purposes.
The methods developed in ANYMOS open up new ways for transport operators to use existing data efficiently and make informed decisions for service planning and operations. The result: a more flexible, attractive, and sustainable public transport offering.