The convergence of artificial intelligence (AI) and urban planning holds significant promise for creating more intelligent, efficient, and sustainable cities. This fusion entails the integration of cutting-edge technologies that can guide decision-making, enhance resource allocation, predict trends, engage citizens, and more. In this framework, where AI is seen as a tool for advancing various urban aspects, there has been a surge in the development of applications, software, and other technological systems tailored to support urban planning. Below, we have highlighted some global studies and technologies applied from urban morphology to community involvement.
Urban Design and Morphology
The concept of the "creature surpassing the creator" has never been closer to reality than when it comes to urban design and artificial intelligence. Recently, a team of urban planners and scientists from Tsinghua University in China developed an AI-based urban planning system capable of outperforming human urban planners in creating projects. In their study, published in the journal Nature Computational Science, the group discusses the factors used to define the ideal urban plan under the "15-minute city" concept. Researchers trained the system using previous human-designed projects and factors believed to be positive characteristics, such as including parks, bike lanes, and entertainment spaces. When testing the system in a 3x3 grid community, they found that the plans generated by the system were as good or even better than those created by experts, and all of this was done in a fraction of a second.
In line with this thinking, Digital Blue Foam, a Singapore-based technology company, has introduced Urban Insights, a tool for rapid urban prototyping. It operates by assimilating data from various open sources, enabling project teams to generate urban scenarios and compare the quality of different locations. DBF collaborates with experts such as Carlos Moreno, the originator of the "15-minute city" concept, and institutions like the Bern University of Applied Sciences to ensure a thorough and scientific analysis of the projects.
As mentioned earlier, using urban prototyping to study different compositions and strategies is likely one of the most promising applications of AI in urban planning. In this context, it's essential to highlight the technology of "digital twins." This AI tool comprises three main components: visualization, prediction, and diagnosis. It provides an opportunity to digitally simulate and test modifications to the urban landscape before implementation, foreseeing how changes in one structure may impact its surroundings. It is a careful approach to urban planning that minimizes stagnation and ensures the city keeps progressing. In practical terms, examples include BuildMedia, which is developing a digital twin of Wellington, New Zealand, and the Chinese company 51World, which has created a digital twin of Shanghai.
Resilience and Climate Adaptation
AI can also assist cities in responding to the impacts of climate change. Predictive modeling, risk assessment, and early warning systems powered by AI contribute to enhancing resilience against extreme weather events.
For instance, Google's Tree Canopy, powered by artificial intelligence, is a tool designed to aid urban planning by mitigating the impact of extreme heat waves. Integrated into Google's Environmental Insights Explorer, it combines AI with aerial images of cities. This enables governments to understand the tree coverage and facilitates more effective planning for new tree plantations.
In addition to this application, it is worth mentioning IBM researchers who have developed a process to accurately identify, map, and measure the amount of carbon that trees in a specific area can store. The study takes into account their species, geometric shape, and leaf volume. For example, the researchers could measure that trees in Manhattan are retaining 52 thousand tons of carbon, pinpointing the specific characteristics of the effectiveness of certain species and where particular types of trees would be most beneficial, aiding in the effective incorporation of vegetation within urban centers.
Infrastructure Planning
Aerial mapping, as seen in the previous example, provides essential information to enhance various urban aspects, making it a valuable tool for data analysis.
In this regard, in 2021, Colombia's National Planning Department (DNP) implemented a pioneering pilot project that combines aerial mapping with urban diagnostics to address the challenge of informal settlements in the country. The project revealed that approximately 85% of the areas identified by algorithmic maps coincided with those verified and outlined by local experts. This high level of accuracy was sufficient to recognize and prioritize areas requiring public policies for space improvement. Based on the promising results of the initial project, the MAIIA software (Automated Mapping of Informal Settlements with Artificial Intelligence) was created. MAIIA is part of the Open Urban Planning Toolbox, a catalog of open-source digital tools for urban planning. The goal is to facilitate the automated detection and mapping of informal settlements in various locations, providing efficient and accessible solutions to housing challenges.
Community Engagement
AI-powered tools can facilitate communication between urban planners and the community. Applications of virtual reality (VR) or augmented reality (AR) can assist residents in visualizing proposed changes and provide feedback, promoting a more participatory planning process.
In this context, it is worth mentioning initiatives that use AI to enhance the monitoring and analysis of data, offering assistance in mobility-related issues such as intelligent traffic management, traffic light control, or camera surveillance. Some initiatives also utilize user-generated data to determine safe routes and monitor public spaces for potential threats to pedestrian safety. In this regard, there are applications specifically focused on urban safety for women that, through artificial intelligence, gather information from users and indicate whether a particular route is considered safe or not.
In the realm of urban planning, a complex and ever-changing field, the potential of artificial intelligence (AI) has garnered attention as a solution to various challenges. However, there are hurdles to overcome before AI can be seamlessly integrated into urban planning. A significant challenge lies in the scarcity of data. The successful application of AI in urban planning relies on accurate and comprehensive data, encompassing factors like population density, traffic flow, and land use. Yet, acquiring such data can be challenging due to privacy issues or incomplete datasets.
However, as AI grapples with challenges related to data scarcity, much discussion has revolved around the role of the citizen, often referred to as the "unpaid data provider." This prompts questions about a situation where we willingly provide information for an urban database, often exploited for profit by private companies. Nevertheless, amid successes and contradictions, one cannot deny the enormous potential of AI in conjunction with urban planning, an avenue that should be increasingly explored. It is crucial, however, to acknowledge the significance of collaboration among urban planners, policymakers, technologists, and the community to ensure that AI solutions are implemented responsibly and inclusively.
