Turing Urban Analytics Programme

The Turing Urban Analytics Programme at Leeds

Turing Fellow Professor Mark Birkin is Programme Director the Turing’s urban analytics programme which is developing data science and AI focused on the process, structure, interactions and evolution of agents, technology and infrastructure within and between cities.

The Urban Analytics programme is framed by four key challenges: urban populations; urban infrastructure; urban environments and urban policy. The programme centres around numerous multidisciplinary research projects.

Below you can read about some of the exciting urban analytics projects taking place at Leeds

You can also learn more about the Turing’s Urban Analytics research projects and upcoming events on the Turing website.


Urban Analytics Digital Twinning Project – Synthetic population estimation and scenario projection

High resolution geographical and sub-population projections are essential for the planning and delivery of services and urban infrastructure developments. SPENSER (a synthetic population estimation and projection model) uses dynamic microsimulation to produce projections under different, user defined scenarios. SPENSER will make high resolution demographic forecasting accessible to stakeholders across a range of application areas, from physical infrastructure planning to health and social care spending, enabling users to run ‘what if’ scenarios and facilitating evidence based planning decisions. This research is led by Turing Fellow Nik Lomax.


Delivering Societal Impact through place-based urban analytics – Act Early: Holme Wood

In January 2020, the Wolfson Centre for Applied Research in Bradford hosted a workshop to focus on Holme Wood. This Turing-led event brought together experts in AI/machine learning, data visualisation and urban analytics with by policy makers and practitioners in areas including health, education, policing, housing, transport, youth services. Critically, the event had substantial representation from front-line practitioners from community organisations together with local charities and faith groups who work in Holme Wood. The workshop addressed two simple questions:

“Can data science improve our collective understanding of a place like Holme Wood?”

“Can we use data to help the community strengthen and grow by ‘acting together’?”

The workshop identified several priority issues that are now being studied using urban analytics. A data extract has been performed allowing data scientists to explore the many interactions between health, education, activity (including travel) and social care. The linked dataset will provide information to underpin a data-driven approach to informing service delivery and present a platform for data science and AI to influence policy decisions. Visualization techniques will be employed to allow stakeholders to understand the potential impact of policy decisions. This exciting programme of research is led by Turing Fellows Mark Mon-Williams and Faisal Mushtaq.


Causal inference and agent-based modelling – Capturing relationships between individuals

Current methodologies lack the sophistication to capture causal relationships between individuals or the resulting feedback that comes about as individuals interact. Agent-based modelling has the ability to simulate individuals, but currently does not accurately capture casual relationships. This project, led by Leeds Turing Fellows Alison Heppenstall and Mark Gilthorpe, connects ongoing work in casual inference modelling to agent-based simulations to robustly capture and simulate causal relationships between individuals.


New data forms for transport policies – Using new types of digital data to support more sustainable travel choices, reducing health, energy, security and other impacts to improve urban lifestyles

Led by Turing Fellows Susan Grant-Muller and Nick Malleson this research analyses novel types of high resolution digital mobility data and other data arising from new technologies, including smartphone and smart city sensors. This will produce new understanding of individuals’ travel choices, enhanced models covering transport, energy, health, security and safety impacts, and ultimately improved policies for the benefit of individuals, communities and the environment.

Tube train

Uncertainty in agent-based models for smart city forecasts – Developing methods that can be used to better understand uncertainty in individual-level models of cities

Individual-level modelling approaches, such as agent-based modelling (ABM), are ideally suited to modelling the behaviour and evolution of social systems. However, there is inevitably a high degree of uncertainty in projections of social systems, so one of the key challenges facing the discipline is the quantification of uncertainty within the outputs of these models. Led by Turing Fellows Nick Malleson, Jonathan Ward and Alison Heppenstall, this project aims to develop methods that can be used to better understand uncertainty in individual-level models. In particular, it will explore and extend the state-of-the-art in two related areas: ensemble modelling and associated emulators for use in individual-level models.


Turing Visualization Research – New visualization methods for profiling datasets and analysis pipelines

Profiling is a backbone of data science, because analysts and researchers need to: (1) investigate data quality, (2) understand the compatibility of different datasets, and (3) check the knock-on consequences of choices made during each stage/iteration of data processing on those that follow.

All three profiling themes present profound challenges, for example, in the area of health, analysts want to use routinely collected health records data but longitudinal analysis is compromised by data quality issues that originate from factors such as changes in recording practice, drift in the use of codes, and government-level policy change. In city analytics it is often essential to integrate heterogeneous data that was originally gathered for different purposes and are, therefore, inherently somewhat incompatible. Both priority themes involve the use of multiple models, with parameters that are sometimes difficult to select or non-trivial to optimise. Choices made for one model can have a profound impact on results further down the data analysis pipeline. This project is led by Turing Fellow Roy Ruddle in collaboration with partners at Newcastle and UCL.


Simulating cities with AI agents – Developing new approaches for modelling human behaviour in cities, for simulating urban dynamics

Human behaviour in urban environments is mediated by a combination of habitual and deliberative decision-making, influenced by prior experiences and current perceptions. The outcome of these decisions are the dynamics of flow and activity we observe in cities every day, and understanding and modelling these behaviours is central to predicting their future evolution. Led by Turing Fellow Ed Manley this project aims to advance the integration of reinforcement learning and agent-based simulation in improving predictions of future cities.


Modelling the joint effects of temporal, heterogeneous datasets

Joint models for temporal datasets that come from different sources, and are therefore heterogeneous, have great potential in revealing information that is not available in each dataset separately. This research project led by Turing Fellow Jeanine Houwing Duistermaat is developing a framework of functional models which can deal simultaneously with sparse and dense temporal predictors, as well as different time lags in their effect on outcomes. An example where such models are needed is to evaluate and interpret the accumulated effects of pollution and lifestyle on health outcomes.