CCRP partners test methods of open data capture, generation, collation, verification, visualisation and dissemination. Four integrated approaches to data provision are being explored: collation of existing open datasets, computational generation of data through inference, crowdsourcing, and live-streaming. Our collective aim is to stimulate a rapid increase in the volume, variety, accuracy, precision and quality of open spatial data available on stocks, at building level. This is necessary to understand stock composition, performance and dynamic behaviour, to support research and policy making.
The CCRP's aim is to develop an international network of collaboratively maintained Colouring Cities open databases/knowledge exchange platforms providing data and information on national buildings stocks. The programme also offers a friendly, informal, creative, non-competitive and experimental space for multidisciplinary teams to work together on content and code. The programme currently involves researchers from six countries, bringing together expertise in computer science, artificial intelligence, machine learning, urban science, data science, physics, architectural history, environmental science, material science, conservation and heritage, housing, planning and architecture.
Examples of research questions of particular interest to the group include: What kinds of buildings exist in the city - typology, age, use, size etc? How many of each are there and where are they located? How long do different types of building last and why and how can we help buildings adapt in a sustainable and resilient way? How useful can Colouring Cities platforms be in analysing and modelling stocks and/or in tracking performance, reducing energy and waste flows, targeting retrofit, and improving housing quality? How can longitudinal data also be collected and what can this tell us about survival, risk and vulnerability in stocks? Can platforms also double up as disaster management tool and how can we share learning of resilient reconstruction? Can common patterns be found across countries in relation to building form and deprivation, poor health and mortality? How similar or different are stocks across countries? Can generated data, when combined with AI and machine learning approaches,reveal patterns and problems unable to be seen before? Might universal spatiotemporal 'rules' of dynamic behaviour exist for stocks, and if so could these be used in new types of simulation model, to more accurately forecast change in the longer term?
The first country to reproduce Colouring London's open code was Lebanon. The Coloring Beirut platform was set up in 2018 by the The American University of Beirut's Urban Lab working with the National Center for Remote Sensing (CNRS) of the Conseil National de Recherche Scientifique (CNRS-Lebanon).
In early 2020 the University of Bahrain's Urban and Housing Lab began exploratory work on Colouring Bahrain, supported by the Bahrain Authority for Culture and Antiquities. Our partners in Bahrain and Lebanon are also looking to work collaboratively to support other research institutions in the Middle East wishing to develop Colouring Cities platforms.
CCRP partners can be identified by our Colouring Cities logo and listing on this, and the Turing Institute's main webpage. This is the only application of Colouring London's open code endorsed by Turing. A dedicated CCRP website, providing a portal to CCRP partner projects is currently being built.
If you are an academic research institution interested in setting up a Colouring Cities platform and would like information on joining the Colouring Cities Research Programme please contact us at the Alan Turing Institute. Our protocols can be found on GitHub at https://github.com/colouring-london/colouring-london/issues/690.