Colouring Cities Research Programme

“The search for new techniques must be regarded as carried out by the human community as a whole, rather than by individuals” Alan Turing, ‘Intelligent Machinery’ (1948) bbbb

What is the CCRP and why is it needed?

The Colouring Cities Research Programme (CCRP) is a collaborative, open knowledge research initiative, designed to drive a step-change in the availability of spatial data on national building stocks.

The CCRP oversees the development of a global network of free Colouring Cities visualisation platforms, and open databases, which provide microspatial data on building stock composition, operation, and long-term dynamic behaviour.

Data are collected to improve stock quality, efficiency, sustainability and resilience, accelerate the move to net zero, and support United Nation’s Sustainable Development Goals. CCRP platforms are managed by academic institutions within participating countries and are research-led. They are also specifically designed to facilitate co-creation with government, industry, the third sector, and citizens. All code and data are released under open licences. CCRP core platform code, and additional country specific code, are released under a GNU General Public License via GitHub. CCRP building attribute datasets are released via individual international platforms under an Open Data Commons Open Database License, and the CCRP Open Manual released on GitHub under an MIT License.

Building stocks are the most significant socio-cultural and economic resource of cities and their largest capital asset. They are where we spend most of our lives, impacting on our happiness, health, wellbeing and financial security, as well as the quality, safety, efficiency, sustainability and economic vitality of the areas in which we live and work. Buildings and built infrastructure are also responsible for almost 40% of global energy-related carbon emissions and 50% of all extracted materials (UNFCC, 2023).

Though information on building characteristics has always been relevant to housing, planning, building design and construction, conservation and taxation, increased demand for large-scale building attribute datasets over the past several decades has been largely due to the need to reduce emissions and to assess risk to buildings in the context of climate change. However, in many countries spatial, and spatiotemporal data, at building level, is extremely hard to obtain, being often highly fragmented across sectors and disciplines, restricted or charged for, unavailable, unverified, and/or only available in aggregated form. Building attribute data are also not generally standardised across countries with time series data rarely available. This prevents comparative analysis, and limits the potential of AI and ML to provide critical insights into the operation and performance of stocks at national and global scale, and over time.

How is the CCRP managed and funded?

The CCRP works to advance a more efficient, whole-of-society approach to knowledge sharing on buildings and cities, at local, national and global scale, and one that prioritises data ethics. To do this it combines a trusted academic framework and research-led approach, with a reproducible open model and a voluntary, distributed management and contributor network. It then uses the resulting structure to help academic partners within participating countries collect, visualise and share standardised building level attribute data. This in turn reduces data costs and improves data quality and availability. It also allows for new informal multidisciplinary, international research collaborations to be rapidly set up.

Colouring Cities platforms and databases are designed as sustainable, resilient, interoperable and permanent resources. They are also designed to be developed incrementally, at low cost, in line with participating institution’s research goals, and to be slowly enriched by researchers, students, citizens, government, industry and the voluntary sector, year-on-year. All CCRP collaborators contribute time to the CCRP in a way that directly benefits their research as well as the CCRP’s goals. All platforms are independently managed and funded through existing programmes or new internal and/or external grants. Demos are generally funded as part of existing research programmes or through internal grants, often with PhD help, and then used to lever support and funding for small platform/database development teams. Between March 2023 and 2024 funding for Colouring Cities platforms, totalling £1 million, was won, individually, by Indonesia, Canada, Britain, Bahrain, Colombia and Sweden.

Screenshot 2024-07-03 at 18 43 59

The Colouring Cities Research Programme is hosted by The Alan Turing Institute in the UK. The Colouring London prototype was built and tested at The Centre for Advanced Spatial Analysis at University College between 2014 and 2019. Turing oversees the CCRP GitHub core code repository and new code development, the CCRP open manual, the CCRP impact study showcase, CCRP protocols and ethical framework, onboarding of academic partners, and chairing of the CCRP International Academic Steering Group. Turing also tests new platform features as part of Colouring Britain’s development, in collaboration with Loughborough University, and with help from the CCRP’s UK regional university network. The CCRP structure is shown above.

CCRP strategy is developed in close collaboration with the CCRP Global Region leads, all of whom are members of the CCRP Academic Steering Group. Hubs have been initiated, for North America by Concordia University (Canada), for Europe by the Leibniz Institute of Ecological Urban and Regional Development Research Data Centre (Germany), for the Middle East by the University of Bahrain, for Latin America by St Andrew’s University and the Universidad Distrital - Francisco Jose de Caldas (Colombia), for Africa by Dedan Kimathi University of Technology (Kenya) and for the Asia Pacific region by the University of New South Wales (Australia) and King’s College London. Hub leads help drive network expansion and knowledge sharing at regional level and chair bimonthly meetings. Platform content is also advised by CCRP expert groups, which focus on specific CCRP data categories, and by national research bodies including Oak Ridge National Laboratory in the US in relation to long-term research goals. Academic partners also benefit from monthly CCRP international software engineering meetings, run by Concordia University, support in platform repository set and free workshops and talks.

Screenshot 2024-06-26 at 11 53 36

Global Region Hubs contain all Colouring Cities platforms within specific regions and run regular CCRP meetings to support platform development and aid new collaborations. Each is managed independently by the university or university consortia leading in each country. Each participating country is set up with a Colouring platform GitHub repository, by Turing, and provided with introductory guidance on using the CCRP’s open source code. Academic leads agree (informally) to work towards CCRP goals, follow CCRP protocols, codes of conduct and branding and to set up a demo platform with a longer term view to national database management. Initial demos for a city or town are recommended to be carried out as part of existing research programmes, working with PhD research software engineers wherever possible. Platforms have been specifically designed to be able to be managed at low cost alongside other core university resources. Where these are embedded as core research resources, and used and contributed to by more than one universities, at an early stage, they are likely to be more sustainable and successful than those that rely wholly on one off grants. However, once demos have been built and embedded, they can then be used to help lever research grants.

All Colouring platforms/databases are run by researchers who require building attribute for their research work. Collaborations between planning, energy and or built environment, and computer science faculties, (across or within universities) are strongly recommended, as is the development of a national academic partner network interested in helping with data moderation and upload and stakeholder engagement at regional level. This allows Colouring platform academic lead/s to level funding to manage the technical side of national databases and collaborate with relevant CCRP Global Region Hubs, and for moderation and upload, analysis and application to be contributed to by national network partners.

The CCRP also offers a welcoming, stimulating and inclusive space for researchers interested in open knowledge and open data systems, in co-working on complex urban problem solving, and in thoughtful, considerate, research collaboration.

Who’s involved?

The Colouring Cities model has been built and successfully tested since 2016 with the help of over 100 researchers worldwide. It is now rapidly expanding, with research institutions in over 30 countries at various stages of engagement. Academic ollaborators are involved in platform set up and management, core code and new feature development, application and testing of data and data capture methods, providing expert advice on data classes and data ethics, and submitting joint funding applications and papers.

Screenshot 2024-07-01 at 16 15 24

Current academic network contributors include, from Australia, University of New South Wales; Austria, AIT and BOKU; Bahrain, University of Bahrain; Britain, Alan Turing Institute and Loughborough University and researchers from the universities of Birmingham, Bristol, Newcastle, Exeter, Manchester, Oxford, Cambridge and UCL; Canada, Concordia University; China/Hong Kong, Hong Kong PolyU; Colombia, Universidad di San Jose; Germany (IOER); Greece, University of Patras, and National Technical University of Athens; Indonesia, King’s College London and Institut Teknologi, Bandung; Kenya, Dedan Kimathi University of Technology; Lebanon, American University of Beirut; Philippines, Ateneo de Manila University; Poland, Poznan University of Technology;Singapore, National University of Singapore; Switzerland EPFL Lausanne; South Africa, Tishwane University of Technology; Spain, IREC; Sweden, Mälardalen University; Turkey, Istanbul Technical University, and the United States, City University New York, Perelman School of Medicine, Penn University and Oak Ridge National Laboratory.

Multidisciplinary expertise within the CCRP includes: computer science, data science, software engineering, urban science, industrial ecology, urban morphology, physics, environmental science, material science, climate change, resilience and risk, building construction, engineering, conservation, housing, health, deprivation, humanitarian aid, planning, architecture, history, architectural history, graphic design, colour theory, open data systems, AI, (including ML and CV), synthetic data, procedural modelling/3D simulation, probabilistic modelling, GIS and spatial data visualisation and uncertainty quantification. Data ethics and ethical use of building level footprints and the identification of security and privacy concerns relating to the visualisation and release of building attribute data form key areas of CCRP work.

Live platform links

All platforms are different stages of development, as shown on the map at the top of the page.

Live links are currently available for:

How do Colouring platforms work

Colouring Cities platforms collect, collate, verify, visualise and release microspatial building attribute data, on building characteristics, building performance, and the short and long-term dynamic behaviour of building stocks. Four main methods of data capture and upload to platforms are used: Open bulk uploads of data uploaded manually and moderated by academia; crowdsourcing at building level with vectorised footprints able to be ‘coloured in’ with information, live; streaming of official data using APIs, and computational methods including AI, ML, GIS based spatial analysis and simulation. Data accuracy is maximised through the ongoing testing and improvement of feedback loops between the methods, verification tools, provision of information on collection method used, and uncertainty quantification (planned). This multipronged approach is also necessary to encourage knowledge contributions from diverse sectors, disciplines and communities.

Platforms are designed to ask basic questions such as How many buildings do we have? What type of building are they? What are they used for? And made of? How old are they? How repairable, adaptable and extendable? How energy efficient are they? How suitable for retrofit and have they been retrofitted? What is their green context? How many demolitions have taken place each site? What is their relationship to the street? Who built them? What is the ownership type? How well do local communities think they work, and what impact does their presence or loss have on communities, and on their physical landscapes, in environmental, economic and socio-cultural terms.

Over 150 classes of standardised spatial data are integrated. These are contained within the 12 CCRP categories: Building location (and ID), Land Use (and economic activity), Building Type & Form, Size, Construction and Materials, Age & History, Street Context (including green context), Planning Context and Protection, Energy Performance, Resilience/Live disaster management, and Community value. Platforms also visualise more standard GIS layers such as flooding zones, planning zones, protected views designated housing/economic/creative areas. Vectorised historical maps are also included, where such maps are available, as are specialised tools such as the CCRP disaster management tool which allows building damage at building level to be mapped live.

Geolocated building footprints operate as mini filing cabinets able to capture, collate, verify, release, and visualise the spatial data, as well as providing information on the building ‘s shape and size. Since 2024 CCRP international expert groups have begun to be formed for all 12 CCRP main data categories. These recommend changes and additions to CCRP core data classes which are then reflected in the CCRP core repository.

CCCP partners use identical interfaces and logos and core data classes to maximise interoperability of systems, to maintain programme and platform quality, and to maximise user trust. No other applications of Colouring Cities open code are endorsed by the CCRP or The Alan Turing Institute other than those specified on the CCRP GitHub site and the Alan Turing Institute ‘s CCRP webpage. This is important as experimentation with CCRP code is actively encouraged. Clear visual branding is also necessary to allow the purpose, principles, and quality of CCRP platforms to be instantly understood regardless of which country a Colouring Cities platform is operating in. Where platforms differ is in additional subcategory inclusion specific to national or regional contexts.

Screenshot 2024-07-01 at 12 02 36

CCRP mission statement

To effect a step change in the range and scale of spatial building attribute data freely available at building level, relevant to socio-cultural, economic and environmental research, urban management and sustainable building practice worldwide;
To provide data necessary to improve the quality, sustainability, efficiency and resilience of stocks, to support and accelerate progress towards United Nations’ Sustainable Development Goals and its New Urban Agenda;
To address data and knowledge fragmentation in relation to building stocks and issues such as data fragmentation, incompleteness, quality, inaccessibility, reliability, aggregation, inconsistency, security and privacy. To do this through mechanisms that facilitate knowledge and data sharing across sectors, disciplines and communities;
To test and integrate a range of data capture methods, including computational approaches; crowdsourcing and live streaming, and feedback loops between these, to support interdisciplinary work and engagement of diverse stakeholders in knoweldge exchange and to maximise data richness, accuracy, granularity, range and geographic cover;
To advance discussion on ethical issues relating to the capture and release of building attribute data, and promote ethical, inclusive, reproducible, and collaborative data science, as laid out in The Turing Way. *
To provide the scale and quality of spatiotemporal data required to facilitate the use AI and machine learning, and the development of digital twins and simulations, to accelerate the move to net zero and to increase resilience in stocks and better understand the stock as a dynamic system; to present access to case studiesshowing diverse applications
To test a decentralised academic governance model for trustworthy, research-led, networked, low-cost, open data platforms providing standardised at building level data across cities and countries.
To reduce research costs, facilitate rapid international research institution collaborations, and drive knoweldge, data and resource sharing;

Contact

If you are an academic institution involved in research into stock sustainability and resilience, and would like to discuss setting up a CCRP platform, please contact Polly Hudson at The Alan Turing Institute at phudson@turing.ac.uk. ___________________________