Data Science and Technology
The Data Science and Technology theme applies new digital technologies to answer research questions and inform decision-making in transport planning, operations, and policy. The theme brings together experts in data science, transport modelling, and the use of broadly-defined artificial intelligence techniques to address challenges in the transport sector. We harness new forms of digital data to generate policy-relevant insights for transport, and closely linked sectors that include energy, health, safety and society. We focus on the full pathway from raw data to operational decisions, including strengthening the evidence base, improving transparency, and building tools that can be used in practice.
Our Methods and Approach
Working across data-driven disciplines, we use a broad suite of analytical and technological methods and combine transport domain knowledge and modern data practice:
• Reproducible research workflows that make assumptions explicit and results traceable through code, data and documentation
• Open tool building that package methods into useable interfaces and pipelines, so that analysis can inform real investment and operational choices
• Behaviourally grounded modelling and simulation where prediction and explanation remain connected to how people and systems actually behave
• Applied AI and decision support that prioritise reliability, interpretability and responsible use when methods are deployed in high-consequence contexts.
Pillars of Inquiry and Impact
• Gen-AI Enabled Data Analysis: using generative AI to accelerate transport analysis by integrating it into established and emerging data-science workflows in practical and accountable ways that reduce analytical friction while preserving scientific rigour.
• Open Data Exploitation: developing methods and tools that unlock values from open datasets and digital infrastructure for transport planning and research.
• Decision-Support Tools: Developing data-driven evidence that converts raw and open data into practical decision-support tools, example includes Propensity to Cycle Tool to inform cycling investment and enable strategic planning for coherent cycle network.
• Digital Twin and AI-Driven Optimisation in Transport Systems: integrating real-time, data-rich virtual models with machine-learning-based predictive analytics to optimise transport routing, scheduling and system-level performance and to support real-time traffic management and operational decision-making.
Current Research Projects
- High-fidelity and high generalisation models of human behaviour
- Network Planning Tools for Scotland
- Propensity to Cycle Tool
- TransiT (Twinning for Decarbonising Transport)
News
Contact us
Email: k.j.butterworth@leeds.ac.uk