Urban Generative Design: The End Game for Smart Cities

Subscribe To Download This Insight

By Dominique Bonte | 3Q 2019 | IN-5594

The way cities are designed and upgraded is about to fundamentally change due to the emergence of urban generative design. The smart cities conundrum has so far mainly been dominated by the Internet of Things (IoT) paradigm, attaching connected sensors to the built urban environment including buildings, road infrastructure, and utilities, allowing optimized management, preventive maintenance, and other types of efficiency improvements and cost savings. However, adding an IoT platform to Operational Technology (OT) infrastructure by definition only represents an incremental step toward managing existing urban assets more efficiently.

Registered users can unlock up to five pieces of premium content each month.

Log in or register to unlock this Insight.

 

Urban Designers to Adopt Generative Design Principles from Manufacturing Industry

NEWS


The way cities are designed and upgraded is about to fundamentally change due to the emergence of urban generative design. The smart cities conundrum has so far mainly been dominated by the Internet of Things (IoT) paradigm, attaching connected sensors to the built urban environment including buildings, road infrastructure, and utilities, allowing optimized management, preventive maintenance, and other types of efficiency improvements and cost savings. However, adding an IoT platform to Operational Technology (OT) infrastructure by definition only represents an incremental step toward managing existing urban assets more efficiently.

Something far bigger is now appearing on the horizon in the form of the automated, generative design of urban environments, both brownfield and greenfield. In fact, it is an illusion to believe that adding just a shallow layer of IoT technology to legacy urban environments will allow cities to address the urban challenges of the future, ranging from the provision of sustainable energy to the adoption of smart mobility and the construction of resilient cities.

While the very concept of urban generative design is only just emerging, leading vendors like Dassault Systèmesand Siemens are already exploring how to leverage their industrial generative product suites for automated city design. Designing new urban areas from scratch, whether they are to be inserted into existing urban contexts (“city in a city”) or as greenfield city designs, like New Clark City in the Philippines, is the only way to fundamentally solve urban issues with design instead of some afterthought plug-in technology.

The idea of urban generative design is borrowed from the manufacturing industry. Often, but not always, linked to the new possibilities offered by additive manufacturing, automated, Artificial Intelligence (AI)-based generative design tools are now starting to be embraced by leading manufacturing players. This allows for the design of geometries and shapes with lower weights at similar strength levels. More generally, more complex objects can be designed, according to any set of design principles and objectives, without having to take constraints linked to manufacturing methods into account. In many cases, generative design tools create multiple variants, the most optimal of which must be manually selected, sometimes based on additional tests on physical prototypes (the design of vehicle bodies, in which the final design is selected based on wind tunnel tests for minimum air resistance, is a good example of this).

So, how can this be leveraged to design entire cities? Cities are essentially aggregated combinations of many types of modules, including buildings, utilities, road infrastructure, campuses, (air)ports, malls, and urban spaces for new forms of shared (micro)mobility. Optimizing the performance of the capacity and flow of energy, goods, people, and vehicles within entire urban zones is extremely complex and requires advanced, AI-based generative design tools to optimize energy demand-response management. How many, which types/shapes of buildings, and how they are clustered together are critical for this optimization. For example, an optimum combination of residential and commercial buildings allows leveling out peak energy demands. The stacking of buildings maximizes exposure to solar energy by minimizing shade. Designing public places for optimized pedestrian traffic, including in the case of emergency evacuations, is another example of how AI can be applied in automated designs.

While buildings are the obvious starting point for urban generative design, the same principles can be applied to road infrastructure design for optimized mobility, the location of pick-up and drop-off locations for shared smart mobility, the distribution of charging stations for electric vehicles, and the design of grids and utilities.

What is unique here is not the generative design of individual components, but the design of the aggregation and juxtaposition of multiple modules and subsystems, akin to the very nature of the smart cities paradigm. In a way, this represents a sort of uber-generative design, or the generative design of generative designs. Urban generative design is closely linked to and enabled by advanced smart cities, digital twins, and urban modeling tools. Urban modeling and generative design very much represent the end game of the smart cities journey.

Dassault and Siemens Leading the Way

IMPACT


As part of the Datavironment Hackathon Dassault Systèmesorganized at the Design in the Age of Experience event during the Milan Design Week in April 2019, one team used the collaborative capacities of the 3DEXPERIENCE platform (CATIA, SolidWorks) to design modular building extensions. Automatically generated bar and restaurant extensions in the form of a city-level template allows for optimized instantiation of extensions throughout the city. xGenerative Design prepares building locations based on city data and the requirements for surrounding building and shadow projection. Metallic parts for the modular structure are created with xDesign with glass panels design based on xShape. xGenerative Design was further used for volume adaptation and randomized positioning of glass panels and structural parts. Another team designed a 2 km long promenade based on shadow, temperature, and pedestrian flux analysis.

As one of the leaders in industrial digital twins and generative design technology, Siemens’ simulation tools can be used to evaluate and test many scenarios, such as the effects new buildings would have on a neighborhood’s air flow conditions. Generative design tools allow a range of designs based on requirements related to strength and functionality to be automatically created and quickly evaluated through simulations. Importantly, the design software can be gradually improved through deep learning processes.

A New Era of Urban Planning and Design

RECOMMENDATIONS


While much of the past focus has been on enhancing existing urban environments with IoT technologies to achieve cost savings and efficiencies, the improvements obtained remain mostly incremental. In the future, entire neighborhoods or even complete cities will be designed from scratch using urban generative design tools. Modeling cities and optimizing operations through digital twins is great; designing them from scratch with AI tools is better. Similar to industrial contexts, modeling and simulation tools can pave the way for the adoption of fully automated generative design, in this case for entire neighborhoods or even entire greenfield city developments, optimizing the geometry, location, and configuration of buildings, energy generation, and road infrastructure for maximum efficiency and comfort. Generative design tools will also be instrumental in creating the circular cities of the future, as explained in ABI Insight A Future of Green and Circular Cities (IN-5576). The inherent complexity of cities demands the use of generative design tools. It is not a matter of choice, but of necessity. It is the only way forward for an increasingly urbanized global environment.

However, there is another important twist to this story. While it is great to generate automated city designs that optimize current urban realities and requirements, the built environment will have to cope with changing needs throughout its long lifecycle. Hence, it will be important to adopt modular designs that allow for the reconfiguration of streets and buildings. Office buildings are already designed for flexible compartmentalization, especially in shared offices and co-working environments, and modular street concepts are already being explored. This possibility of flexibly modifying the urban built environment represents a shift from approaches based on one-off designs followed by a long life cycle of managing static assets to an ongoing and continuous design effort, within certain constraints, based on reconfiguring both the physical and Information and Communication Technology (ICT) environments in terms of short- and long-term shifts in needs and requirements. This will disrupt and transform not only the construction industry but also the transportation and energy industries in terms of the use of new materials and building concepts.

Services

Companies Mentioned