A new software to predict and control urban expansion by 'training' AI. It is an online platform capable of reading maps and transforming them into models for sustainable urban planning. The new resources provided by IBM and SEforALL to identify the needs of a community and better plan its future.
What better stage than COP29 in Baku to launch new applications capable of helping cities effectively control their urban expansion and energy needs, to avoid falling into the dangerous trap of “urban sprawl.”
The launch is led by IBM and the independent organization SEforALL, Sustainable Energy for All, which, with pro-bono funding from the IBM Sustainability Accelerator, have developed two AI-based projects that will assist policymakers in sustainable urban development.
Mapping Urbanization and Identifying Energy Needs
The starting point is developing regions, where urban expansion often occurs in an uncontrolled manner, neglecting the needs of the population and the issues triggered by climate change.
By leveraging Artificial Intelligence (AI), the organizations have developed management software and an online platform to create a current and future vision of the built environment and urban realities in regions of Africa and India.
“At Sustainable Energy for All (SEforALL), we believe that integrating artificial intelligence into energy sector planning and assessments, particularly for developing countries, will significantly contribute to designing comprehensive solutions for many of the development challenges currently faced by the Global South and its population,” emphasizes Damilola Ogunbiyi, CEO and Special Representative of the UN Secretary-General for Sustainable Energy.
Buildings Becoming More “Readable”
Initially, the focus will be on expanding the online platform Open Building Insights. This tool consolidates data visually from a map, providing information about buildings in countries facing major urban planning challenges. Thanks to AI integration, decision-makers, as well as non-expert users, can deduce the height of buildings, their footprint area, and their usage type directly from the map. In this way, all stakeholders can make informed decisions about urban expansion, implementing sustainable development.
The interactive map of Open Building consolidates models created by the German Aerospace Center (DLR), which estimates building heights through Open Energy Maps. By simply observing the map, it is possible to understand how much energy a specific building might consume, based on its size and its use (residential or non-residential).
The platform is freely available and, to date, has already been used by Kenya to improve its energy planning, anticipating the needs of millions of citizens.
Modeling Urban Growth
The second tool presented at COP29 by IBM and SEforALL is Modelling Urban Growth (MUG). MUG is an open-source artificial intelligence model designed to predict where and how cities will expand. It is trained using historical satellite imagery, geographic data (such as slope and altitude), demographic information, and structural data (such as road networks), all integrated into a time series.
MUG helps users map future urbanization and identify infrastructure needs, supporting decision-makers in prioritizing communities and regions that require infrastructural interventions, electrification, and general energy services. This open-source model is publicly available on GitHub.
Currently, the model is trained with data from selected developing regions, but the most interesting aspect is that it can be adapted and retrained by users for any country in the world, using publicly available data.
