Predictive Data Analytics for Urban Dynamics
Lead Research Organisation:
University of Leeds
Department Name: Sch of Geography
Abstract
Identifying the factors that encourage and discourage attendance in urban spaces is vital from both academic and practitioner perspectives. For policy makers, an understanding of what attracts people to city centres, and what discourages them, is vital for urban planning and emergency management. From an academic perspective, understanding the drivers of footfall is essential in order to answer questions of mobility, inclusivity, and accessibility of opportunities.
Simply quantifying footfall, let alone dissecting the underlying drivers, is extremely challenging. Although there are a series of diverse 'big' datasets that are emerging which can contribute to footfall estimates, none in isolation provide a comprehensive picture. In addition, there are no standard methods that are appropriate for assimilating dynamic, diverse, noisy, and biased data sources to create a complete picture.
To address these gaps in our understanding of urban dynamics, this project will embark on an ambitious programme of methodological development and empirical data analysis. It will adapt relevant methods from fields such as computer science (e.g. machine learning and artificial intelligence), atmospheric modelling (e.g. data assimilation and ensemble modelling), and geography (e.g. GIS and spatial analysis) to create a robust model of footfall. Indicative data source that will underpin the analysis include: footfall data collected by Leeds City Council CCTV cameras; Census workday population estimates; dynamic weather data; geo-located Twitter data; times of public events and holidays; business opening hours; Wi-Fi sensor footfall data; and others as they become available. There is great potential to leverage these data as both an explanatory tool for understanding what has been driving city-centre attendance and as a predictive tool for forecasting future footfall under different scenarios. Leeds City Council are actively involved in designing the research questions, identifying data sources, and will jointly supervise the research.
This project will feed into work across the CDRC remit, providing baseline and temporally nuanced urban populations, along with an explanation of their drivers. It will feed into real-world urban management systems at Leeds City Council, as well as broader cities and future cities literature. Finally, it will act as a foundation for future work in emergency planning and diurnal population movements, as well as longer-term predictions of city accessibility.
Simply quantifying footfall, let alone dissecting the underlying drivers, is extremely challenging. Although there are a series of diverse 'big' datasets that are emerging which can contribute to footfall estimates, none in isolation provide a comprehensive picture. In addition, there are no standard methods that are appropriate for assimilating dynamic, diverse, noisy, and biased data sources to create a complete picture.
To address these gaps in our understanding of urban dynamics, this project will embark on an ambitious programme of methodological development and empirical data analysis. It will adapt relevant methods from fields such as computer science (e.g. machine learning and artificial intelligence), atmospheric modelling (e.g. data assimilation and ensemble modelling), and geography (e.g. GIS and spatial analysis) to create a robust model of footfall. Indicative data source that will underpin the analysis include: footfall data collected by Leeds City Council CCTV cameras; Census workday population estimates; dynamic weather data; geo-located Twitter data; times of public events and holidays; business opening hours; Wi-Fi sensor footfall data; and others as they become available. There is great potential to leverage these data as both an explanatory tool for understanding what has been driving city-centre attendance and as a predictive tool for forecasting future footfall under different scenarios. Leeds City Council are actively involved in designing the research questions, identifying data sources, and will jointly supervise the research.
This project will feed into work across the CDRC remit, providing baseline and temporally nuanced urban populations, along with an explanation of their drivers. It will feed into real-world urban management systems at Leeds City Council, as well as broader cities and future cities literature. Finally, it will act as a foundation for future work in emergency planning and diurnal population movements, as well as longer-term predictions of city accessibility.
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
ES/R501062/1 | 30/09/2017 | 29/09/2021 | |||
1944290 | Studentship | ES/R501062/1 | 30/09/2017 | 31/12/2021 | Annabel Elizabeth Whipp |
ES/P000401/1 | 30/09/2017 | 29/09/2024 | |||
1944290 | Studentship | ES/P000401/1 | 30/09/2017 | 31/12/2021 | Annabel Elizabeth Whipp |
Description | The findings of my work have been used to inform decisions regarding the location of footfall cameras which were to be installed by Leeds City Council. The footfall cameras and the data that they produce are utilised to inform retail decision and to better understand the movement of the ambient population. The findings were of particular benefit to gathering data about the night-time economy and it's users; however developments of this have been put on hold due to Covid19. |
First Year Of Impact | 2020 |
Sector | Government, Democracy and Justice |
Impact Types | Cultural,Societal,Economic,Policy & public services |
Description | GISRUK funding for the ten best postgraduate abstract submissions in 2019 |
Amount | £200 (GBP) |
Organisation | GISRUK |
Sector | Charity/Non Profit |
Start | 03/2019 |
End | 04/2019 |
Description | PhD summer school for advanced spatial modelling |
Amount | £300 (GBP) |
Organisation | University College London |
Sector | Academic/University |
Country | United Kingdom |
Start | 08/2019 |
End | 09/2019 |
Description | Predictive Data Analytics for Urban Dynamics |
Amount | £81,612 (GBP) |
Funding ID | 1944290 |
Organisation | Economic and Social Research Council |
Sector | Public |
Country | United Kingdom |
Start | 09/2017 |
End | 09/2021 |
Description | Data partnership with Leeds City Council |
Organisation | Leeds City Council |
Country | United Kingdom |
Sector | Public |
PI Contribution | The main contributions that I made to the partnership were my analytical and programming skilled which enabled detail analysis and visualisation of the data. Leeds City Council did not have researchers with these skills previously, which meant that the full potential of the data they were collecting had never been explored. |
Collaborator Contribution | The data was provided by Leeds City Council and I was given the opportunity to work alongside the City Centre Management team and the Head of Economic Development. Working with the team at Leeds City Council enabled me to explore the data they had access to and to address issues which concerned them, for example the relationship between footfall counts and the night-time economy. |
Impact | From the work conducted in partnership with Leeds City Council, a decision was made by Leeds City Council to review their relationship with the third party collecting the data due to issues flagged during the period of data analysis. Leeds City Council decided, following the presentation of my research to a cross-team board, that footfall camera data held vast potential and therefore should be invested in financially. |
Start Year | 2018 |
Description | Research open day for smart cities research and data open to members of the public |
Form Of Engagement Activity | Participation in an open day or visit at my research institution |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Public/other audiences |
Results and Impact | Over 100 members of the general public and individuals from third sector organisations attended a presentation of smart cities research and the importance of the use of big data. Feedback from the event highlighted that as people gained an awareness of how their data can be used by researchers for public good, they became more willing to volunteer their own data in the future. This has significant impacts for a wide range of research areas which rely upon individual level data. |
Year(s) Of Engagement Activity | 2019 |
Description | Seminar on AI and machine learning |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Undergraduate students |
Results and Impact | Approximately 59 students and members of the public attended a talk which was run by myself on AI and machine learning and the benefits this can bring to issues which concern the general public. |
Year(s) Of Engagement Activity | 2020 |
Description | Seminar on digital twins |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Undergraduate students |
Results and Impact | 45 students, business professionals and members of the general public attended a talk I gave on digital twins and how they can be utilised for predictive analytics in direct relation to my project. It increased public awareness of the benefits which can be provided by this research. |
Year(s) Of Engagement Activity | 2019 |