Data-Smart City Solutions

Search

By Laura Adler

In cities across the world, governments use data—about infrastructure, health and safety, or citizen satisfaction—to improve services. But data also has a critical role in shaping the very space of the city, informing the design of new buildings, infrastructure, and neighborhoods. With the help of big data and analytics, urban planners can now use simulations to anticipate the impact of urban development programs. Using these tools, cities can become more sustainable and strategic, while the planning processes become ever more inclusive. 

The most fundamental benefit of simulation is the ability to mitigate the problem of “unintended consequences” by using realistic models to predict effects on land valuation, employment patterns, and transportation mode choice. In urban planning, optimizing one system often comes at the expense of functionality in other areas: the construction of much-needed housing can lead to overburdening the local transportation infrastructure; campaigns for water conservation can, ironically, damage a city’s water infrastructure.

Simulation allows planners to anticipate cascading effects across urban systems from water management to energy and waste management to parking. Simulations are an important bridge between theory and experiment: without affecting real-world situations, designers can predict the outcome of interventions across a range of scenario specifications. Simulations have thus become an indispensable tool for urban planners.

Simulations driven by big data can also help to ameliorate long-standing democratic deficits in the planning process. Making decisions about the future of cities has always relied on in-person interaction, as officials solicit the input of citizens in the context of city hall hearings or community planning meetings. Such meetings historically exclude many residents: parents who are taking care of their children cannot attend an evening meeting; working people cannot attend during the work day; and citizens with limited English language capabilities might have difficulty participating without translators. But a range of new tools are allowing urban planners to expand the scope of their input by drawing in data on urban systems and allowing residents to voice their ideas online.

Simulations have been used for many years, with land use models and economic forecasting serving a central role in urban planning since at least the 1970s. But these models were hampered by limited technologies: at the time, data collection was difficult and costly, computational power was severely limited, analytic capabilities were rudimentary, and computer-based visualization still lay in the future. Planners used statistical models to estimate the relationship between factors like population growth and urban density, but these simulations were still a far cry from integrated analytical tools and vehicles for public outreach and engagement.

Much has changed in recent years. Today, online data collection and sensor systems in the environment are generating unprecedented quantities of data, enhancing the predictive power of statistical models. Spatial information systems are becoming more advanced, allowing for better geographical analysis and compelling visualizations that can communicate potential outcomes to citizens. And the growth—by leaps and bounds—of computing power allows planners to easily calculate the interrelationships of multiple urban systems. On the basis of these transformations, platforms for urban simulation have proliferated, including both comprehensive systems and system-specific prediction tools.

Comprehensive Urban Simulation Platforms

A number of new tools have emerged to help urban planning researchers, practitioners, and residents envision the impacts of urban transformation. UrbanSim, founded by University of California, Berkeley Professor Paul Waddell, allows users to run simulations, draw from a library of open data, and produce visualizations. The program, which is free and open source as part of the Urban Data Science Toolkit, is designed to help planners understand the diverse impacts that might be expected from new forms of street design, mixed use zoning, or policies to promote urban density. In addition to open source code, UrbanSim is gradually developing a library of datasets from the cities that use the system, which are accessible to other users.

UrbanSim aims to help urban planners better understand the impact of interventions, but also to increase the inclusiveness and transparency of planning processes by involving citizens throughout. Planners can share scenarios with citizens in visually compelling ways, but residents can also generate their own designs and share their vision with neighbors and government officials.

Also led by academic researchers, MIT’s Changing Places initiative has developed CityScope, an urban simulation tool that integrates physical representation—using Legos—with projections and visualization tools. Using sophisticated analytics, users can see the impacts to be expected from shifting density by manipulating the Legos, or by setting different rules for infrastructure and mobility systems. The in-person visualization tool thus helps planners and stakeholders predict, quantify, and visualize the outcome of urban transformation. The program, which is still evolving, has been used to help officials and their constituents envision the effects of change from Riyadh to Hamburg to Cambridge’s Kendall Square.

With Participatory Chinatown, Emerson College offered a more hands-on approach to community engagement through digital simulation. The program used a multiplayer game format to engage citizens in a number of simulated neighborhood activities inside a digital recreation of Boston’s Chinatown. Acting as assigned characters and as themselves, players were asked to explore a number of activities within the game —finding a job, a place to live, or a place to socialize—and use the experience to generate urban planning priorities to guide city officials.

Governments are also developing their own tools to help public officials access, analyze, and interpret urban data. Australia’s “urban intelligence network,” or AURIN, is a state-run resource for the nation’s cities and towns that provides datasets and online tools for analysis, modeling, and visualization. The program aims to support “evidence-based policy and decision-making” in cities and towns across the country. By centralizing data and analytics tools, AURIN ensures that individual cities can limit redundant efforts and better learn from one another’s experiences.

Simulating Specific Urban Systems

While a small number of governments and universities are tackling comprehensive simulation, a wide range of tools are being developed to model specific urban systems. The original urban planning simulation, Land Use Transport interaction, or LUTi models, estimate population, land use, and transportation parameters for urban development over the long term, often 30 to 50 years. As explained in an overview from the Centre for Advanced Spatial Analysis, LUTi models were used in the 1960s and 70s but their usefulness was limited by data quality and computing power.

With the rise of GIS in the 1990s, these models became more informative and data has only increased from there. Smartphones are producing a growing pool of unstructured data embedded with geographic information, which planners can start to integrate into the predictive models. And cities are beginning to use these tools to help citizens understand the impact of planned or proposed developments. In Chicago, the Metropolitan Planning Council developed the TOD Calculator (for Transit Oriented Development) to help residents learn about the economic and sustainability benefits of potential TOD sites.

A number of tools have emerged that focus specifically on sustainability, including water and energy systems. Phoenix WaterSim was developed by Arizona State University researchers to help the Phoenix metropolitan government estimate supply and demand in order to effectively manage its limited resources. WaterSim emphasizes the role of visualization in helping to make data intelligible and accessible to stakeholders and decision-makers.

To address water management in a very different climate, the Zofnass Information Tool was developed in Chelsea, MA, to help citizens understand opportunities for water management improvements—like green roofs and porous pavement—and estimate the environmental benefits from these interventions. 

In the area of energy, both urban and building design are critical, and new programs can help designers and planners predict energy demand and identify conservation opportunities. Umi is an urban modeling platform run by MIT’s Sustainable Design Lab, which estimates the environmental performance of buildings and cities with modules for embodied energy, walkability, and daylight and shading. A similar program from the École Polytechnique Fédérale de Lausanne, CitySim, allows designers to estimate energy demand and plan for the integration of renewable energy sources.

This is just the start. As evidenced by a 2015 conference at MIT—the Computers in Urban Planning and Urban Management conference—researchers around the globe are hard at work modeling diverse urban systems to meet different city challenges. By bringing these researchers together, conferences like this one help to produce collaborative networks of researchers who can better design opportunities for integration across simulation platforms.

Despite Advances, Challenges Remain

Although data, analytic capabilities, and accessibility are rapidly expanding, there are important challenges that urban planners, designers, and stakeholders still face. As in other areas of big data, the first of these is the need for systems that can manage and integrate diverse data, pertaining to distinct but interconnected urban systems. Collecting, cleaning, managing, and integrating varied forms of data remains a major challenge.

More fundamentally, the question of what goes into simulation models—and what gets left out—has important consequences. What might go wrong if a model used for planning residential zones fails to account for the historical legacy and continued reproduction of housing segregation? How might the exclusion of renewable energy sources from transportation planning simulations skew the sense of what is possible for sustainable public transit? Just as important as what we do once we run the simulations are the choices we make when we build them: ensuring that the models underlying these informative experiments represent a balanced and unbiased set of assumptions.

What goes into the simulation model? An example from UrbanSim

Source: UrbanSim

 

Finally, in order for simulations to truly drive inclusion in the planning process, designers must find ways to make software feel useable to a non-technical audience. Ideally, stakeholders should be able to develop their own simulations without in-depth knowledge about data sources and statistical models, so that they can develop their own understanding of what is possible in their city. Only with simple, accessible simulation programs can citizens become active generators of their own urban visions, not just passive recipients of options laid out by government officials.

This article has been corrected to reflect that UrbanSim is not a project of Autodesk.

About the Author

Laura Adler

Laura Adler is a PhD student in Sociology at Harvard. She received a Bachelors from Yale University and a Masters in City Planning from the University of California, Berkeley. Laura's research interests include urban planning and social policy in the US and abroad, with recent academic work focused on the relationship between urban governance and technology. Prior to beginning graduate study at Harvard, Laura worked for the City of New York's Department of Information Technology, where she focused on long-term technology strategy in support of the city's operations and expanding broadband access for New York City residents.

Connect

Comments

comments powered by Disqus