MULTIMODAL urban transport: integrated modelling and simulation towards net-zero, inclusive mobility
Lead Research Organisation:
Loughborough University
Department Name: Architecture, Building and Civil Eng
Abstract
Domestic transport is the UK's highest emission sector, and congestion in cities is costly (e.g. London £5.1bn in 2021). Drastically reducing urban car dominance is imperative to reach the UK's 2050 net-zero target, but also an unparalleled opportunity to create more equitable, inclusive and accessible cities of the future across the country. Recent UK investments of approximately £15bn seek to radically transform urban mobility and modality: £2bn for half of urban journeys to be cycled/walked by 2030 (e.g., cycle lanes, mini-Holland schemes), £5.7bn City Region Sustainable Transport Settlements (e.g., Manchester bus and cycle schemes), and £7bn to level up local bus services.
To realise full investment potential, and develop holistic adoption pathways towards net-zero, inclusive mobility, multimodal transport must be effectively planned, managed and operated, with people and their differences as a core consideration. This is challenging for a complex system-of-systems. On the supply side, modes compete for limited road space on shared infrastructure, creating conflicts. On the demand side, modes complement each other in intermodal journeys, jointly influencing uptake. For example, cycle lanes promote cycling, but may impact road speeds and exacerbate congestion and pollution, highlighting the need to evaluate person-level mobility and system-level emissions. A recent survey reported two-thirds of disabled respondents finding cycling easier than walking, highlighting the need to consider the broad disability spectrum and the potential for cycle lanes to improve access for all. Therefore, holistically optimising cycle lane schemes, as with all multimodal schemes, requires integrated methodologies: fully capturing multimodal transport systems' distributed and interconnected processes, the complexities of modal competition and complementarity, and the heterogeneity of traffic and population.
My research will overcome these research challenges and develop the first multiscale digital twin for the transport-people-emission nexus using a truly integrated approach to model and simulate multimodal urban transport, advancing and coalescing my adventurous research in multimodality, using traffic flow theory, agent-based modelling, and machine learning. This will enable the development of holistic adoption pathways towards net-zero, inclusive mobility through scenario testing and optimisation, with guidance and recommendations to support implementation. Leading a strong consortium of 3 cities and 12 partners, covering the entire multimodal transport value chain, I will collaboratively exploit the digital twin to realise UK strategic agendas: net-zero; Equity, Diversity and Inclusivity (EDI); and levelling-up. By holistically enhancing mobility for everyone, my Fellowship also will propel the Green Revolution for economic growth, leveraging the net-zero mission to unlock new business opportunities, and establish the UK as a global leader in digital technologies to tackle climate change. I will deliver a strong positive impact on making net-zero a net win for people, industry, the UK, and the planet, thereby enabling both me and the UK to become world leaders in multimodal urban transport, at the forefront of research and innovation.
To realise full investment potential, and develop holistic adoption pathways towards net-zero, inclusive mobility, multimodal transport must be effectively planned, managed and operated, with people and their differences as a core consideration. This is challenging for a complex system-of-systems. On the supply side, modes compete for limited road space on shared infrastructure, creating conflicts. On the demand side, modes complement each other in intermodal journeys, jointly influencing uptake. For example, cycle lanes promote cycling, but may impact road speeds and exacerbate congestion and pollution, highlighting the need to evaluate person-level mobility and system-level emissions. A recent survey reported two-thirds of disabled respondents finding cycling easier than walking, highlighting the need to consider the broad disability spectrum and the potential for cycle lanes to improve access for all. Therefore, holistically optimising cycle lane schemes, as with all multimodal schemes, requires integrated methodologies: fully capturing multimodal transport systems' distributed and interconnected processes, the complexities of modal competition and complementarity, and the heterogeneity of traffic and population.
My research will overcome these research challenges and develop the first multiscale digital twin for the transport-people-emission nexus using a truly integrated approach to model and simulate multimodal urban transport, advancing and coalescing my adventurous research in multimodality, using traffic flow theory, agent-based modelling, and machine learning. This will enable the development of holistic adoption pathways towards net-zero, inclusive mobility through scenario testing and optimisation, with guidance and recommendations to support implementation. Leading a strong consortium of 3 cities and 12 partners, covering the entire multimodal transport value chain, I will collaboratively exploit the digital twin to realise UK strategic agendas: net-zero; Equity, Diversity and Inclusivity (EDI); and levelling-up. By holistically enhancing mobility for everyone, my Fellowship also will propel the Green Revolution for economic growth, leveraging the net-zero mission to unlock new business opportunities, and establish the UK as a global leader in digital technologies to tackle climate change. I will deliver a strong positive impact on making net-zero a net win for people, industry, the UK, and the planet, thereby enabling both me and the UK to become world leaders in multimodal urban transport, at the forefront of research and innovation.
Organisations
- Loughborough University (Lead Research Organisation)
- Lime Technology Ltd (Collaboration)
- Department of Transport (Collaboration)
- NOTTINGHAM CITY COUNCIL (Collaboration, Project Partner)
- UNIVERSITY OF CAMBRIDGE (Collaboration)
- Transport for Greater Manchester (Collaboration, Project Partner)
- TRANSPORT FOR LONDON (Collaboration, Project Partner)
- Arup Group (Collaboration)
- Connected Places Catapult (Project Partner)
- Vectare Limited (Project Partner)
- Veitch Lister Consulting (UK) (Project Partner)
- Department for Transport (Project Partner)
- Arup Group (Project Partner)
- Atkins (Project Partner)
- Sustrans (Project Partner)
- Lime Technology Limited (Project Partner)
- PTV System Software und Consulting GmbH (Project Partner)
- German Aerospace Center (DLR) (Project Partner)
- O2 Telefonica Europe plc (Project Partner)
- Immense Simulations (Project Partner)
Publications

Davies R
(2024)
A Multiscale Framework for Capturing Oscillation Dynamics of Autonomous Vehicles in Data-Driven Car-Following Models
in IEEE Transactions on Intelligent Transportation Systems

Rowan D
(2025)
A systematic review of machine learning-based microscopic traffic flow models and simulations
in Communications in Transportation Research

Wei Y
(2024)
A Memory-augmented Conditional Neural Process model for traffic prediction
in Knowledge-Based Systems
Description | - Understanding how vehicles interact is key to managing traffic. New AI-based models can predict traffic flow more accurately than traditional methods, but their strengths and weaknesses are not well studied. Our review highlights what's missing and suggests future directions to make these AI models more useful, reliable, and widely applicable. - Current AI models can mimic how cars follow each other but miss wider traffic patterns like traffic jams. Our new MultiscaleCF model combines real driving data with traffic science to better capture these effects. It makes traffic simulations more realistic, improving accuracy of vehicle behaviour and traffic flow by up to 42%. - We introduced MemCNP, the first AI model of its kind for predicting traffic patterns, even when data is limited. By learning from past traffic and using memory to handle complex situations, it predicts future traffic more accurately. MemCNP works for cars and bikes, supporting smarter, greener transport planning. - Traffic prediction is vital for smart transport systems but struggles with sudden changes in traffic patterns. To tackle this, we introduce ST-Align, a new AI model that uses memory to adjust to shifting traffic conditions over time and space. Tests on real data show ST-Align makes more accurate and reliable predictions. |
Exploitation Route | Our research advances AI-based traffic modelling and prediction, offering more accurate and adaptable tools for traffic management. By identifying gaps in current AI models and introducing new methods (MultiscaleCF, MemCNP, and ST-Align), these outcomes can help transport authorities, planners, and researchers improve traffic simulations, design smarter transport systems, and manage congestion more effectively. Others can build on these models to develop AI-driven solutions for real-time traffic control, autonomous vehicle systems, and sustainable urban mobility, as well as extend them to other transport modes and cities. |
Sectors | Digital/Communication/Information Technologies (including Software) Transport |
Description | We developed TraffEase to tackle a pressing global challenge-cities worldwide struggle with congestion, emissions, and pollution, yet the power of transport data remains largely untapped. In London alone, congestion costs £3.85 billion annually, transport accounts for 28% of the UK's carbon emissions, and air pollution causes 43,000 premature deaths each year. Despite cities investing heavily in data collection, we noticed that the complexity of transport models and a shortage of skilled analysts-affecting over 90% of UK city councils-hinder real impact. We saw an urgent need for an AI-powered solution to makes transport data accessible, actionable, and impactful. TraffEase is set to transform transport management by simplifying complex transport data for smarter human decision-making, ultimately reshaping urban mobility and creating more sustainable cities worldwide. With AI-powered insights, authorities can implement targeted interventions-optimising bus routes, adjusting congestion charges, enhancing cycling infrastructure, and streamlining public transport schedules to reduce delays and overcrowding. This will improve passenger experience, maximise infrastructure investment, and create a safer, more reliable, and accessible transport network that meets community needs while driving local economic growth. For the public, this translates to less time in congestions, fewer disruptions, seamless connections, cleaner air, and healthier cities. |
First Year Of Impact | 2025 |
Sector | Communities and Social Services/Policy,Digital/Communication/Information Technologies (including Software),Transport |
Impact Types | Societal Policy & public services |
Description | Integrated National Transport Strategy Roadshow |
Geographic Reach | National |
Policy Influence Type | Contribution to a national consultation/review |
Impact | The Integrating Transport: Have Your Say, Leicester regional roadshow gathered public feedback to shape future transport plans. The consultation informed policy and strategic priorities by highlighting local needs and challenges, ensuring that emerging transport strategies aligned with community expectations. It also refined specific projects and schemes, as feedback on issues like public transport reliability and cycling safety led to adjustments in project design and prioritisation. |
Description | Policy workshop: AI Opportunities Action Plan - lessons and opportunities for the public sector |
Geographic Reach | National |
Policy Influence Type | Contribution to new or improved professional practice |
URL | https://www.lboro.ac.uk/policy-unit/news/2025/traice-ai-action-plan/ |
Description | Manchester Prize (finalist award) |
Amount | £1,000,000 (GBP) |
Organisation | Department for Science, Innovation and Technology |
Sector | Public |
Country | United Kingdom |
Start | 04/2024 |
End | 03/2025 |
Description | E-scooter routing |
Organisation | Lime Technology Ltd |
Country | United Kingdom |
Sector | Private |
PI Contribution | Models and analysis |
Collaborator Contribution | Data and feedback |
Impact | Research publications |
Start Year | 2022 |
Description | Establishment of European Association of Activity-Based Modeling |
Organisation | Arup Group |
Country | United Kingdom |
Sector | Private |
PI Contribution | Research and leadership |
Collaborator Contribution | Serving as steering committee members |
Impact | Establishment of the association; fosters collaboration between practitioners and academics by hosting dedicated workshops at various conferences and venues; support knowledge building as part of academic curricula and continuing education and enables cooperation among university lecturers and practitioners. |
Start Year | 2024 |
Description | Establishment of European Association of Activity-Based Modeling |
Organisation | Department of Transport |
Department | DfT, DVLA and VOSA |
Country | United Kingdom |
Sector | Public |
PI Contribution | Research and leadership |
Collaborator Contribution | Serving as steering committee members |
Impact | Establishment of the association; fosters collaboration between practitioners and academics by hosting dedicated workshops at various conferences and venues; support knowledge building as part of academic curricula and continuing education and enables cooperation among university lecturers and practitioners. |
Start Year | 2024 |
Description | Impact of cycle lanes on buses |
Organisation | Transport for London |
Country | United Kingdom |
Sector | Public |
PI Contribution | Models and analysis |
Collaborator Contribution | Data and feedback |
Impact | Research publication |
Start Year | 2024 |
Description | Iterative testing and deployment of TraffEase prototype |
Organisation | Nottingham City Council |
Country | United Kingdom |
Sector | Public |
PI Contribution | Technology and prototype |
Collaborator Contribution | Time and capacity for iterative testing and feedback |
Impact | A test-deployed TraffEase prototype; a policy brief on 'Application of Generative AI in the Public Sector' |
Start Year | 2024 |
Description | Iterative testing and deployment of TraffEase prototype |
Organisation | Transport for Greater Manchester |
Country | United Kingdom |
Sector | Public |
PI Contribution | Technology and prototype |
Collaborator Contribution | Time and capacity for iterative testing and feedback |
Impact | A test-deployed TraffEase prototype; a policy brief on 'Application of Generative AI in the Public Sector' |
Start Year | 2024 |
Description | Iterative testing and deployment of TraffEase prototype |
Organisation | University of Cambridge |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Technology and prototype |
Collaborator Contribution | Time and capacity for iterative testing and feedback |
Impact | A test-deployed TraffEase prototype; a policy brief on 'Application of Generative AI in the Public Sector' |
Start Year | 2024 |
Title | TraffEase |
Description | TraffEase is an AI assistant transforming transport management by simplifying complex transport data for smarter human decision-making. By unlocking the power of transport data with purpose-built AI agents, TraffEase enables users to ask transport-related questions in plain English and instantly access the most advanced models for analysis. Powered by cutting-edge deep learning, TraffEase uncovers hidden relationships within datasets, providing precise, actionable insights. Imagine an AI-powered team working alongside your organisation-delivering accurate transport insights to drive smarter, faster decisions for everyone. TraffEase empowers decision-makers to implement targeted interventions that optimise mobility, reduce congestion, and cut emissions. |
Type Of Technology | Software |
Year Produced | 2025 |
Impact | TraffEase is set to transform transport management by simplifying complex transport data for smarter human decision-making, ultimately reshaping urban mobility and creating more sustainable cities worldwide. With AI-powered insights, authorities can implement targeted interventions-optimising bus routes, adjusting congestion charges, enhancing cycling infrastructure, and streamlining public transport schedules to reduce delays and overcrowding. This will improve passenger experience, maximise infrastructure investment, and create a safer, more reliable, and accessible transport network that meets community needs while driving local economic growth. For the public, this translates to less time in congestions, fewer disruptions, seamless connections, cleaner air, and healthier cities. |
URL | https://www.lboro.ac.uk/schools/science/news/2024/ai-transport-project-shortlisted/ |
Company Name | Transhumanity |
Description | Transhumanity develops TraffEase, a software platform which provides predictive traffic analytics with the aim of facilitating more sustainable travel planning decisions. |
Year Established | 2024 |
Impact | TraffEase is set to transform transport management by simplifying complex transport data for smarter human decision-making, ultimately reshaping urban mobility and creating more sustainable cities worldwide. With AI-powered insights, authorities can implement targeted interventions-optimising bus routes, adjusting congestion charges, enhancing cycling infrastructure, and streamlining public transport schedules to reduce delays and overcrowding. This will improve passenger experience, maximise infrastructure investment, and create a safer, more reliable, and accessible transport network that meets community needs while driving local economic growth. For the public, this translates to less time in congestions, fewer disruptions, seamless connections, cleaner air, and healthier cities. |
Description | Establishment of European Association of Activity-Based Modeling |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Establishment of the association; fosters collaboration between practitioners and academics by hosting dedicated workshops at various conferences and venues; support knowledge building as part of academic curricula and continuing education and enables cooperation among university lecturers and practitioners. |
Year(s) Of Engagement Activity | 2024,2025 |
URL | https://eaabm.org/ |
Description | Keynote speaker in Transport AI Conference |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Industry/Business |
Results and Impact | Transport AI UK will provide a one-stop shop for understanding the opportunities and challenges of AI: - a platform for attendees to gain insights on which AI-powered products, services and tools can deliver now to help local authorities and consultants - the AI development curve: how to use AI to add value by time-saving, cost-saving and automating routine tasks - a look ahead to coming innovation: planning an AI readiness strategy - understanding of the resourcing, procurement, human and ethical factors that will underpin the development and take-up of AI across the transport sector |
Year(s) Of Engagement Activity | 2025 |
URL | https://www.transportai.uk/conference-2025 |