Project Summary
UrbanTide, in collaboration with Edinburgh University is developing the uMove_AictiveRoutes tool. This will use unique and innovative AI and advanced analytics to reveal travel behaviour patterns among commuters and leisure cyclists in Edinburgh, offering insights into journey purposes and route selection. This solution addresses the lack of understanding of route purpose and selection for active travel behaviours, which hinders effective infrastructure planning. By integrating data from multiple sources, including the Cycling Scotland Open Data Portal, Edinburgh Universities timestamped UniCycle bike hire data, and Sustrans impact reports, the solution aims to showcase how new innovative network analysis and AI exploitation techniques can provide accurate insights on travel behaviours to inform decisions that reduce car dependency, congestion, and pollution. Ultimately, this supports progress toward active travel goals, improves user experience, and contributes to climate targets.
Project Achievements
Using novel data sources and AI techniques, we addressed key gaps in understanding active travel, particularly journey purpose and route choice. By building a transport network model of Edinburgh, we identified decision points and conflict zones using unique analysis, network cleansing and leveraging mobile data to improve mode detection accuracy. Within the project timeframe we advanced our technology readiness, generated new IP, and demonstrated that global datasets can offer broader product potential than localised cycle hire data. We also engaged with key stakeholders to explore how our insights can drive greatest impact and how integrating our innovation with existing data and workflows enhances the value of our uMove solution.
Conclusions
Collaboration was key to this project’s success, enabling data sharing and the testing of scalable AI and analytics methods. Our flexible approach led to the discovery that global datasets offer more accurate and valuable insights than localised sources, particularly for understanding cycling behaviour at scale. This allowed us to address critical gaps in knowledge around active travel—specifically journey purpose and route selection—which are essential for planning effective infrastructure. These insights can help local authorities overcome existing challenges in analysing travel behaviours and support efforts to reduce car dependency, congestion, and emissions. By providing a better understanding of active travel behaviours and journey patterns, our solution can support planning and policy decisions. We intend to continue developing this work, providing local authorities and other stakeholders with vital intelligence on movement and transport usage.
Next Steps
This project demonstrated the feasibility of our unique methodology with strong engagement from stakeholders, laying the groundwork for real-world application. As a next step, we plan to enhance our existing
uMove data insights platform by integrating global datasets and scaling our network creation methodology. We aim to validate this approach across different geographies and contexts, strengthening its applicability for
active travel planning and analysis.