The ‘Breathing City’ project is demonstrating how the use of open data can create a digital version of a city, i.e. a digital twin, to improve the wellbeing and safety of urban populations and provide valuable insights to educate society on the impact of pollution
Understanding how our cities are operating in real time can play an important role in helping us to solve some of society’s biggest challenges. The ‘Breathing City’ project – an initiative led by Slingshot Simulations – aims to demonstrate how open data can be used to create a digital representation of a city, which can provide valuable insights about the impact of pollution, and help decision makers to improve the wellbeing and safety of city populations.
Back in March 2020, the Lloyd’s Register Foundation and the ODI offered a stimulus fund to help projects to increase access to data and drive innovation in the engineering sector – with an emphasis on improving safety, and this is one of them.
Open innovation – where a business or organisation ‘opens up’ their innovation process to gain knowledge and ideas from external sources – can enable new solutions to address social, economic and environmental challenges, as we have outlined in our manifesto for sharing engineering data. With support from Lloyd’s Register Foundation, we have been enabling organisations to create and test new solutions to existing problems.
‘The Breathing Cities project has provided Slingshot Simulations with a unique opportunity to showcase the feasibility of creating a digital twin from primarily open data sources that can reliably act as the foundation for urban data science projects for universities, local authorities, and local businesses. We anticipate replicating this across cities in the UK and further afield in the coming 12 months’ – David McKee, Slingshot Simulations
Key facts and figures
- Slingshot Simulations has created a digital twin of Leeds city centre, using multiple open data sources, which can be used to understand the challenges faced around air pollution in the city.
- The digital twin enables city decision makers to simulate different scenarios in which changes to city infrastructure could have a significant impact on the effects of air pollution.
- The digital twin has the potential to be rolled out across different cities, with a reusable simulation being produced as part of the project. Newcastle-under-Lyme council have approached the team to explore replicating this in Staffordshire.
- Slingshot’s work has created a foundation for helping to city planners in a COVID-19 world, which will be further explored through their new follow-on project – A Living Model for People and Place – which is funded by Innovate UK and supported by the University of Leeds VirtuoCity Centre for City Simulation, Leeds City Council and Newcastle-under-Lyme Borough Council.
What was the challenge?
Air pollution is one of the leading contributors to increased mortality around the world, with established links to numerous heart and respiratory diseases. Roughly 7 million deaths a year are rooted in air pollution and more than 80% of people who are affected by poor air quality live in urban areas, such as cities. City decision makers must do more to mitigate poor air quality and protect citizens, by making better decisions around city planning.
The consequences of planning decisions on air pollution are not currently well understood
Areas around Leeds have the worst air pollution outside of London. Changes need to be made to improve air quality in and around the city, but the consequences of planning decisions on air pollution are not currently well understood.
City decision makers need to be able to make decisions about city planning, without fearing that those decisions will have a negative impact on citizens, or the wider city ecosystem. To do this, they need access to information about how different scenarios might unfold, and what the consequences of those scenarios might be.
City decision makers need to be able to make decisions about city planning, without fearing that those decisions will have a negative impact on citizens, or the wider city ecosystem
One of the core examples looks at traffic flows around Leeds city centre. Main roads that run through the city centre such as Neville Street are often heavily congested, which leads to a much higher concentration of air pollutants being produced in that particular area. Given the high amount of regular footfall in this area, in part because the street is so close to the main train station in Leeds, this level of pollution could be causing negative health effects in a large number of people.
Being able to redirect some of this traffic, by optimising the way that traffic flows or even creating new routes for drivers to take, could have a significant positive impact on the air quality in this area.
Having a solution that scales across different areas would help greatly in addressing this
Leeds is not the only city in the UK that faces air quality and pollution issues; many cities are facing a similar problem. Having a solution that scales across different areas would help greatly in addressing this.
How are they solving the problem?
Slingshot has created an interactive digital twin of Leeds city centre, by integrating a variety of open datasets into a cloud-based real-time simulation platform. The model can be used to simulate different scenarios which recreate citizen interactions with heavily polluted areas. These ‘what-if’ scenarios can help city decision makers to determine the best course of action for preventing citizen exposure to air pollution, without actively affecting the safety of individuals.
Slingshot has created an interactive digital twin of Leeds city centre, by integrating a variety of open datasets into a cloud-based real-time simulation platform
Through creating this solution, Slingshot aims to bring air quality and population health to the centre of the smart city discussion, and demonstrate how digital twins are a great tool for providing evidence that decision makers can use to understand the impact of potential changes.
The team underwent four phases of delivery in order to build the digital twin of Leeds. This included building the model on open geospatial data; integrating pollution and traffic data; integrating footfall data; and testing the model based on a set of what-if questions.
During the first phase, the team focused on building up the static base layer mapping of the particular areas of focus in Leeds. These areas include Neville Street, which is the most polluted street outside of London, and the immediate area around Leeds train station. Open data integrated here comes from sources such as Open Street Maps, the NASA Shuttle Radar Topography Mission and European Space Agencies Sentinel Programmes. The model generates a 3D model from open datasets in a roughly 1 mile radius. They have also been able to run this at up to a 30 mile radius, equivalent to 2,500 square km. The data used provides the basis for a similar process to be undertaken in other areas, both within Leeds and other cities.
The next step was to start integrating traffic and pollution data into the model. This included Automatic Number Plate Recognition combined with pollution data from the UK Air Information Resource and the National Atmospheric Emissions Inventory. Then the team worked to integrate pedestrian footfall data from National Rail and Leeds City Council.
Creating a digital twin of Leeds required the team to build up a progressively more complex model
The final phase of the project investigated specific scenario analysis of what if situations, including prototyping people moving in a confined space to take account of social distancing. The results of these simulations will be made publicly available as datasets and as animations.
Although Slingshot’s simulation platform was developed prior to this project, creating a digital twin of Leeds required the team to build up a progressively more complex model, in order to make the simulated scenarios accurate and useful to users of the model.
Data was key as it allowed the model to demonstrate an increasing array of different complex scenarios
The simulations for Leeds generate roughly 10GB/s of data and take roughly 20s to run a simulation equivalent to eight hours of traffic, modelling every 5s of agent behaviour. This is far more efficient than any other method available to city decision makers in Leeds.
Data was key as it allowed the model to demonstrate an increasing array of different complex scenarios. By working with stakeholders to source and use different open data, Slingshot has provided the basis for similar models to scale up in different locations.
What was the impact of taking this approach?
Slingshot has demonstrated the value of taking a new approach to understanding the challenges and risks of air pollution and how it can be combated. The insights generated should inform and support Leeds City Council to meet their targets to reduce the effects of air pollution and improve planning for pedestrianisation.
Learning and simulations from this project will be applicable and extensible to other urban areas within and beyond Leeds, helping to understand the impact of air pollution in urban environments. The use of NASA Satellite data within the project means there is scope to scale internationally.
The project has provided Slingshot with an opportunity to produce a working digital twin of a major UK city, which helps to address a city priority around the reduction of the harmful effects of air pollution. Throughout the project, the Slingshot team has made new connections with local community groups, businesses and national bodies, such as the Centre for Digital Built Britain and the Digital Twin Consortium, which convene around the digital twin agenda.
Learning and simulations from this project will be applicable and extensible to other urban areas within and beyond Leeds, helping to understand the impact of air pollution in urban environments
The project has also laid the foundations for Slingshot to secure further funding to continue their work, as part of Innovate UK’s ‘Business-led innovation in response to global disruption’ competition. This opportunity will allow Slingshot to work with Leeds City Council to explore how transport and citizens will move around the city post COVID-19.
What lessons did they learn?
The major challenge that Slingshot faced was integrating some of the open datasets into the model, as the datasets often conformed to different standards. Openly sourced data can sometimes be quite difficult to validate and it is not always clear until you access the raw data as to what state it will be in. Bus routes data is an example of this; sometimes the entry refers to a specific bus stop, whereas other times it can refer to a broader location.
Key tips and advice
Engaging with the wider city community as early as possible was seen to be a major learning from this project. The impact of COVID-19 somewhat limited the ability to do this from the start of the project, but Slingshot found that once they started to engage with people, there was more interest than they thought there would be. For example, they found that citizens were particularly interested in understanding the effects of 5G technology on air pollution, and used the digital twin model to address some of the fears around this, through community engagement activities.
Slingshot has demonstrated that with the right datasets available, modelling areas of a city is entirely possible
Slingshot has demonstrated that with the right datasets available, modelling areas of a city is entirely possible. However, it could be difficult to replicate the scenarios created by this model in other cities if similar datasets do not exist for those areas, as we have found through our ‘Scaling data innovation’ research. Access to the relevant, good quality data will play an important role in scaling the Slingshot model to other areas.
Slingshot will continue to work with Leeds City Council on a follow on project focusing on providing up to date information on traffic and people flow as we return to work following the pandemic. By using existing data to forecast and then iteratively, in real-time, update those forecasts, Slingshot wants to support councils by helping them address the pandemic from multiple environmental and economic angles.
There are a number of other city councils who have expressed an interest in adapting the model for their context. Newcastle-under-Lyme council have approached Slingshot to explore replicating this in Staffordshire, and a number of other organisations have asked for a similar approach, but with additional features built in.
View Slingshot Simulation’s latest story: Can Digital Twins Sustainably Save Money? Yes it can
Our other stimulus fund projects
Blog, Case study 06/08/2020
Case study: Unlocking data on brownfield sites
- Built environment and housing
Blog, Case study 06/08/2020
Case study: Creating a standard to help people find construction knowledge
- Built environment and housing
- Data infrastructure