Data Science and the Built Environment

The special collection can be accessed through the ASCE Library at: (https://ascelibrary.org/jupddm/Data_Science_Built).

I am pleased to announce this special collection of the Journal of Urban Planning and Development, dealing with data science and the built environment. The papers present research on how cities develop their intelligence, how the new technologies shape the lived experience of urban residents and communities, and how they inform urban planning, design and development research, and practice. The collection includes six papers written by eleven authors, representing institutions from China, New Zealand, the UK, and the United States. They discuss methodological, application, and implementation issues related to new sources of urban data and informatics.

The impetus for this special collection was to capture this moment of major transformation in the sources, types, formats, and volume of data used to understand and intervene in the built environment. Since the mid-1980s, advances in geospatial technologies, digital data capture, and decision support systems have offered useful tools to urban planners and design professionals for storing, analyzing, and visualizing information. However, the pace of technological progress has been faster than scholars' and professionals' ability to fully explore, adapt, and exploit the technologies and tools to meaningfully contribute to practice. Their employment in planning and place-making practices has proven challenging, owing primarily to organizational adoption issues (Nedović-Budić and Godschalk 1996; Nedovic-Budic and Pinto 2001; Vonk et al. 2005) but also to their reliance on institutional datasets, most of which are general in content and have been collected at coarse spatiotemporal resolutions. As the technology has moved from geographic information systems (GIS) and WebGIS, to spatial data infrastructures (SDIs) and geoportals, user-generated content enabled by Web 2.0, social media, and urban sensors, informational capacities have been ever expanding, albeit with very little time allowed to understand the impacts. Like their predecessors, the recent developments in volunteered geographic information (VGI, Goodchild 2007), crowd-sourcing facilities, social media, and sensors (Resch et al. 2015; Corcoran 2017) have opened up new possibilities. They promise to enable a further step into the democratization of data-dependent processes and have blurred the boundary between data users and data producers to create a hybrid category of data produsers (Budhathoki et al. 2008, 2010). The most recent technological wave is likely to complement structured databases with real-time and fine-resolution data streams. This would allow for an innovative, creative and, most importantly, immediately applicable combination of information as a basis for urban planning, design, and decisionmaking. Beyond the exclusively engineering- and computing-based smart city paradigm enabled by the Internet of Things (IoT, Rathore et al. 2016), the most valuable aspect of integrating data from disparate sources—traditional (institutional or structured) and those embedded in sensors and individualized devices—is to enable both the factual (objective) and perceptual (subjective) understanding of the urban environments and communities, their characteristics, experiences, needs, and aspirations.

The second impetus for this special collection is to affirm the expansion from geographic information science, whose body of knowledge and multidisciplinary foundation have by now solidified (UCGIS 2016), to data science. This has happened somewhat spontaneously, with the introduction of the smart city paradigm, and sources of data multiplying and diversifying, going beyond formal institutional collections (e.g., census, county business patterns, local government records, and maps) and beyond structured and systematic process (e.g., VGI, big data). Much of the early research on GIS and SDIs drew on various sources of disciplinary knowledge, from mathematics to politics, but, to a large extent, remained isolated (Nedovic-Budic 2000; Nedović-Budić and Budhathoki 2006; Budhathoki and Nedovic-Budic 2007; Nedović-Budić et al. 2011). With the emergence of new technologies, data sources, and applications (Web 2.0, social media, and sensors), scholars and professionals across many disciplines have engaged in and contributed to the introduction of data science frameworks. More direct engagement and interaction with other scientific communities, building on the achievements of information science, decision support systems, computer engineering, operational research, and cognitive, management, and organizational science, has brought about data science as the common and overarching interdisciplinary reference.

However, as things change, they also stay the same. Many of the original concerns related to the increasing capacity to capture and integrate data from many different sources have persisted and possibly accelerated; for example, the turning of technology into surveillance and bureaucracy tools, misuse of information, and privacy issues, as Pickles (1994), Davis (1990), and others warned more than two decades ago. At the same time, new data sources, technologies, and tools are offering thoughtful and constructive insights into cities, and hold the potential to make them better. Recent publications testify to both benefits and shortcomings. Kitchin (2014) and Kitchin and Perng (2016) give us a broad view of open source, big data, code, and data infrastructures, and their implications. Through critical analysis, they examine an emerging “programmable city.” Thakuriah et al. (2017) focus on big data, urban informatics, and applications, while Ran and Nedovic-Budic (2016, 2017, 2018) conceptualize and provide a prototype of a spatial data infrastructure that connects directly with policies through an integrated framework and modeling (spatially integrated policy infrastructure—SIPI). Reflecting on the decades-long effort to build decision support systems that can suit the needs of the planning process, Geertman and Stillwell (2020) offer a comprehensive review of recent scientific advancements in framing, developing, and implementing planning support systems. Underlying and linking the various aspects of data, technology, and cities is the interdisciplinary work of Batty (2017), who examines urban flows, networks, and interactions, and introduces theories and methods that reveal the deep structure of how cities function.

This special collection complements ongoing research efforts and speaks to recent technology and data trends, as well as the opportunities and challenges they present to urban planning, design, and development scholarship and practice. Chen and Yang (2020) integrate big data (WeChat) with other local data sources (e.g., points of interest—POIs) to conduct simulations for understanding the impacts of tourism-related traffic and facility use in a historic quarter in Suzhou, China, they propose alternative designs and transit options that would minimize negative impacts. Maurer (2020) assesses the ontologies of named places tagged in posts of several widely used social media platforms, including Twitter, Instagram, and Foursquare Swarm, and, based on a 36-month sample of 18 million posts, suggests the advantages of names over tagged coordinates in terms of the user experience and economic utility. Nesse and Airt (2020) catalog nine evaluations of Google Street View (GSV) as an alternative observational tool to primary data collection for studying components of urban streetscapes; they explore GSV's general suitability for capturing larger permanent objects, depending on the time the images are recorded and the observers' training and expertise. Using bibliometric analysis, Niu and Silva (2020) take a broad sweep at varied sources of crowd-sourced data, such as social media, point-of-interest data, and collaborative websites, to review data mining, fusion, applications, and use in spatial, sociodemographic, and perception analysis; they discuss the challenges these data sources pose to researchers and practitioners. Poplin (2020) presents and tests a prototype online e-footprints game, an example of a “serious urban planning game,” aimed at collecting data on residents' energy consumption and saving behavior, and gathering feedback on the quality of the game's interface—visualization, aesthetic, and usability. Drawing from the literature on complex systems, theories of sociotechnical interactions and synoptic planning practices based on engagement, and a survey of urban planning stakeholders, Schindler et al. (2020) examine the challenges related to the influence and adoption of spatial decision support tools in planning practices, pointing to the local context as the key issue in tools' development and implementation.

This collection reflects the state of the art as well as the continuous struggle of planning and design scholars and practitioners with data and technology. We have come a long way, but also have a long way to go. In fact, we are in a constant chase to understand and make use of ever-changing and rapidly evolving technologies and tools. Altogether, the intention behind this special collection is to elicit reflections on the potentials and limitations of the new data sources and related methods and applications. While progress and change are inevitable, our task is to make sense of them and take the advantage to deal more responsibly and effectively with built environments and the communities that inhabit them.