How can data help shape the future of cycling in our cities?

The event brought together researchers, policymakers and advocates to explore how new forms of data can inform planning, improve infrastructure and turn evidence into action for safer, more sustainable urban mobility.

With limited financial and spatial resources, deciding how to balance transport priorities is never simple. Expanding cycling infrastructure often involves difficult trade-offs, but data and research can bring much-needed clarity. By using evidence to design, test and evaluate interventions, cities can make decisions that are transparent, equitable and effective. The workshop showcased how both traditional and emerging datasets, from travel surveys to GPS traces, combined with modern methods such as machine learning, can help cities understand cycling behaviour and create conditions where cycling can thrive.

Measuring cycling activity

The workshop was opened by Silke Kaiser (Hertie School), who shared her work on graph neural networks for predicting citywide traffic volumes. Her model, GNNUI, outperforms existing approaches. By accounting for urban specificities, such as complex street networks, heterogeneous street characteristics, and distinguishing between quiet and busy streets, GNNUI GNNUI produces accurate predictions, especially in scenarios where very limited traffic sensor data is available. Such traffic volume estimates provide a foundation for better-informed transport infrastructure planning.

Daniel Velázquez (ISGlobal) examined the health co-benefits of replacing short car trips with cycling across the EU. His scenarios quantify the emissions, health and economic impacts of reducing car use. They show that compact urban planning, improved cycling infrastructure and shifts in travel behaviour can bring major environmental and public health benefits.

From evidence to action: policy insights

Carol Sobral (Hertie School) presented research on Estimating Cycling Mode Share in European Cities, carried out with Mark Nieuwenhuijsen and Lynn Kaack. Mode share, the percentage of trips made by each transport mode, is a key measure of urban mobility. Yet many European cities lack reliable data, as travel surveys are expensive, infrequent and often inconsistent. The  team used machine learning and open data on city characteristics to estimate cycling adoption across Europe.  The results offer cities a realistic starting point for setting cycling targets and open up new opportunities for studying links between cycling, health, air quality, congestion and emissions.

Marcelo Lampkowski (ICLEI Europe) shared insights from the CDP–ICLEI Track platform, an open dataset that collects annual reports from cities on climate, mobility and sustainability actions. The data help track progress and identify where support is most needed. Lampkowski also acknowledged the challenges that cities face, from data gaps and limited funding to the difficulty of implementing policies, while highlighting the growing value of shared data in driving climate and mobility transitions.

New data frontiers for cycling analysis

Thomas Kjær Rasmussen (Technical University of Denmark) presented research on Bicyclist Route Choice Behaviour, using large-scale GPS data from Copenhagen to explore how infrastructure influences cyclists’ route selection.. His work shows the importance of well-connected, continuous cycling corridors and finds that investment in such infrastructure is not only good for mobility but also economically attractive.

Danielle Gatland (HeiGIT) showcased methods for analysing cycling routes and assessing infrastructure using large-scale trajectory and open data. Her work demonstrates how data can identify missing links and network weaknesses, helping cities target improvements that make the biggest difference for riders.

A keynote on Cycling for Sustainable Cities was given by Professor Ralph Buehler (Virginia Tech). His message was clear: safe infrastructure is essential but needs to be supported by a coordinated mix of policies such as land-use planning, parking management, traffic calming, education and public transport integration. He also emphasised the importance of equity, calling for inclusive planning that considers vulnerable and risk-averse groups at every stage.

Across all presentations, one message stood out: data is not only for measurement, it is a catalyst for change. From machine learning models that fill knowledge gaps to international networks that share experience, each contribution showed how evidence-driven approaches can help cities build fairer, more efficient and more sustainable cycling systems. By bringing research and policy closer together, data can turn aspiration into transformation – one ride at a time.

This project has received funding from the European Union's Horizon Europe research and innovation program under Grant Agreement No 101057131, Climate Action To Advance HeaLthY Societies in Europe (CATALYSE).