Connected Cities

Designs on Data: Smart technologies and sophisticated analytical tools are helping us understand and address urban challenges.
[ Page 2 of 2 ]    
Architectural Record
By Katharine Logan

Barcelona’s ability to try out an urban-design concept in limited areas of the city before rolling it out at full scale is, from Singapore’s perspective, something of a luxury. “Singapore is a city-state, and there’s rarely the room for us to experiment with our plans,” according to Siau Yong Ng, director of the Singapore Land Authority’s geospatial division. So, in a global first, the city is synthesizing all of the 3-D efforts of its various government agencies, along with vast stores of associated data from existing geospatial and other platforms, to create a digital twin of the entire city.

The $54 million project, scheduled for phased deployment beginning this year, will provide a single, authoritative digital platform for government and private-sector users, as well as researchers and the public, to better understand the city and to explore ways to improve it. Users will be able to pretest concepts and services that range from energy and food production to mobility, from development proposals to long-term land-use planning.

Astonishingly detailed for such an enormous construct, the platform’s semantic 3-D modeling provides information on terrain attributes, transportation infrastructure, and buildings and their geometry and components, right down to floorplans and materials composition. Multiple sources of static, dynamic, and real-time city data, including information from government agencies, the internet, and Internet of Things devices, enrich the platform with demographics, traffic, and weather. According to a statement from Singapore’s National Research Foundation (NRF), which led the development of the model, “the potential uses of Virtual Singapore in tackling livability issues are limitless.”

For the city’s designers, planners, and decision-makers, the platform allows sharing and reviewing project documentation in context, and to conduct more meaningful public consultations. In turn, this enhanced contextualization and collaboration also allow for a more integrated consideration of how proposed changes will affect the public realm.

For example, planners can simulate the effect of proposed green roofs on temperature and light intensity in the surrounding area, overlay heat and noise maps on existing and proposed developments, or model a building’s influence on wind flows in the street. With slopes, steps, and curbs modeled, it will be possible to visualize universally accessible routes. The platform will also allow users to filter buildings based on pre-set parameters: apartment blocks suitable for solar panels under the country’s Greenprint initiative could be quickly identified by number of stories and roof type.

From their own experience with building information modeling, architects may already be familiar with many of these capabilities, but to bring them together in a project of Virtual Singapore’s scope and scale offers to revolutionize the way cities understand themselves. Concentrating so much information in one, publicly accessible platform, however, is also raising unprecedented issues of privacy and security. “This information will help our daily lives, but it could also fall in the wrong hands and create problems for Singapore,” George Loh, the NRF’s director of programs, told Reuters. “We need to think about that. We need to be two or three steps ahead.”

A newer entrant in the smart-city stakes, Los Angeles, is also using big data to understand risks and opportunities. Projects that address two prominent L.A. priorities include an artificial intelligence-enabled disaster preparation and management program, and a mapping-based community-development initiative.

According to the United States Geological Survey (USGS), odds are that a magnitude 6.7 earthquake will hit Los Angeles in the next 30 years (with lower, but still considerable, odds for a quake of greater magnitude). As part of its efforts to mitigate the impacts of such a disaster, the L.A. Emergency Management Department (EMD) is testing a software platform that incorporates data on hundreds of attributes for three key vectors: the natural environment (from such sources as topographical maps, soil surveys, and USGS seismic sensors); the built environment (building code and permit documentation, zoning maps, satellite imagery); and demographics (from the U.S. Census Bureau). The software combines these data to run simulations of, say, the thousand most likely earthquakes—or particular quakes of defined magnitude, depth, and location—and applies machine learning and predictive analytics to forecast impacts at the census-block level. In the event of an actual disaster, it would provide near real-time information on where and what types of damage to expect and what populations were likely to be affected.


Virtual Singapore, a synthesis of the digital models of various government agencies and vast stores of data, can be put to a wide variety of uses, including producing a visual display of apartment resale values (left), or helping bicycle commuters map the best route between two points (right).

“Which buildings are going to collapse and block emergency routes, so we can prioritize upgrading them? Which major highways in the supply chain are going to collapse, and which will survive, so I can designate them as alternates?” These are some of the questions Aram Sahakian, EMD’s general manager, hopes the software can help with. He is planning to run a simulation in a public-private partnership with several grocery distribution centers located on the San Andreas Fault. Since they are a critical part of the city’s infrastructure, Sahakian wants them up and running three to four weeks after a major quake. He’s hoping that the data-based predictions will motivate the companies to start planning for resilience. “As businesses, it’s in their interest too,” he says. About a year into the trial, it’s too early for Sahakian to judge how useful the software will be, but it’s promising, he says.

While the EMD uses data to prepare for the worst, RegenCities, a 2018 American Planning Association Smart Cities Award–winning research initiative by SOM, aims for the best. RegenCities draws on an SOM-developed, systems-based methodology, which the firm calls “health topography.” It uses geo-referenced information to analyze a city’s vital signs. A scan of public data from open-source portals is combined with census-tract data for a defined area, and categorized according to five urban systems: built, natural, infrastructure, socioeconomic, and cultural. The data are then aggregated, mapped, and cross-­referenced against 10 principles, or traits, of regenerative cities (such as energy, economy, and mobility) to reveal areas of abundance and scarcity. The data sets themselves are not intended to be conclusive, says Gunnar Hand, leader of SOM’s city-design practice. Instead, they help to focus discussion among the people involved.

Beginning with a health-topography assessment conducted for South L.A., SOM partnered with local businesses, nonprofit organizations and community leaders, and the Los Angeles mayor’s office to found a community development initiative called RemakeLA. Returning to data, this time for property ownership, RemakeLA identified and mapped about 5,000 municipally owned surplus sites (property severed during a road construction, for example, and left to deteriorate into a weedy nuisance). It then developed an online geohub to open a public conversation about these sites’ potential for community-based economic development, entrepreneurship, and reinvestment.


SOM’s Los Angeles research project RegenCities collects data from open-source portals, combining them with census-tract data. Deploying a systems-based methodology to understand vulnerabilities and opportunities, it analyzes this information across several indicators, including those for public health and socioeconomic standing.

As a result, SOM is now collaborating with the architecture program at Los Angeles Trade Technical College, whose students come predominantly from South Los Angeles, to develop a pilot program for one of the sites. Participants will use the geohub to identify a parcel suitable for locating an open architecture studio—a base for community-student architectural collaborations. If approved, the facility will be designed and built by students in the college’s programs. One primary goal for the pilot “is to prove to the mayor’s office that permitting use of these parcels could be a valuable tool in their economic-development box—empowering the public,” says Hand.

Responding to the particular circumstances and priorities of their home place, these examples of how cities are using technology to understand and improve themselves vary wildly. Yet they share a common theme: the integration of data to craft a comprehensive strategy. “We need a holistic approach to cities,” says Rueda, “because our main subject is humanity.”

Katharine Logan is an architectural designer and writer focusing on design, sustainability, and well-being.


[ Page 2 of 2 ]    
Originally published in Architectural Record