Spatial Statistics

RIDIR: Scalable Geospatial Analytics for Social Science Research

nsf PI: Sergio Rey & Co-PI: Ran Wei, Amr Magdy, Vassilis Tsotras Oct 1, 2018 - Sept 30, 2021

The project's goals are to enhance the open source spatial analysis library PySAL in four main directions. The first innovation concerns new methods for the treatment of uncertainty in the American Community Survey data. Second, a new framework for statistical inference in the space-time analysis of segregation will be developed. Third, scalable algorithms for regionalization and spatially explicit aggregation of areal units will be produced. Finally, the project will develop a new scalable spatio-temporal data management layer to support the application of these three sets of analytics to large datasets. The software developed here will be included in enhanced (and new) modules in PySAL. This will afford the development of new types of applications using different interfaces from plug-ins for desktop GIS, both open source and proprietary, to the development of software as service systems and the ability of installations where on-premise requirements are binding (Census Research Data Centers). This portfolio of use cases will support a wide range of scientific disciplines and will provide direct benefits to society through new geospatial tools that will improve the state of the art in urban planning, economic development, and spatial public policy.

Comparative Regional Inequality Dynamics: Multiscalar and Multinational Perspectives

nsf PI: Sergio Rey & Co-PI: Yehua Wei May 1, 2018 - Apr 30, 2021

This research project will comparatively analyze the dynamics of regional inequality in the United States and China using recent developments in geographic information science. Because the investigators will focus on core-periphery structure, place mobility, and scalar effects, this project will significantly improve basic understanding of spatial inequality. This project will advance knowledge about the dynamics of regional income inequality in a number of different ways. It will provide a systematic study of the multi-scalar nature of regional inequality, thereby providing a clearer picture of the dynamics of regional inequality and its relationship to scale. It will provide new insights about place mobility and about the core-periphery structure in regional development, and it will provide new perspectives regarding the role of geography in regional inequality. Project findings should prove valuable to domestic and international organizations, such as non-profit organizations and the U.S. and Chinese governments, giving them improved understanding of inequality and facilitating the development of more informed policies that address development issues in equitable ways.

Neighborhoods in Space-Time Contexts

nsf PI: Sergio Rey Sept 1, 2017 - Aug 31, 2020 Learn more

This research project will advance the way in which neighborhoods are defined in urban social science research. The definition of a socioeconomic neighborhood is an important and central issue across many research domains. Neighborhoods serve as the organizational units to frame empirical research that has examined a wide array of issues, including the ability of social networks to produce collective efficacy in a spatial context, the relationships between concentrated poverty traps and violence, spatial sorting and segregation, and the role of neighborhoods as seeds for wider urban economic development. Despite this importance, consensus has not yet emerged regarding how to operationalize neighborhoods in practice. Moreover, the role of spatial structure largely has been ignored in existing approaches to neighborhood definition. This project will produce new methods for neighborhood identification and analysis that draw on recent developments in geographic information science and spatial statistics. These new methods will enhance the urban social science toolkit enabling researchers to carry out empirical work that can be replicated, thereby moving urban policy research onto stronger analytic foundations. The new methods and analytics will be implemented in a publicly available open-source longitudinal neighborhood analysis package. The project will provide training to a post-doctoral research associate and will include a partnership between academia and industry, with applications to commercial market segmentation and analysis.

New Approaches to Spatial Distribution Dynamics

nsf PI: Sergio Rey Jul 31, 2014 - Aug 31, 2018 Learn more

This research project will improve understanding of spatial inequality dynamics through methodological advances in measurement and modeling. Understanding the nature of spatial inequality dynamics is vital to both basic social science and to public policy, yet existing methods and models provide incomplete views of these spatial dynamics. While a central focus of inequality research has been on the evolution of the aggregate income distribution, much less attention has been directed at the spatial pattern of inequality and pattern dynamics. Spatial inequalities can have important implications for social cohesion, economic growth, and the design of policies targeted at reducing the level of inequality. The advances produced in the project will have wide applicability. In addition to spatial income inequality dynamics, many other social and economic phenomena have distributions that evolve in space and time. Software packages will be delivered as open-source projects and accompanied with extensive tutorials and documentation to facilitate broad dissemination across the social sciences.

Network Effects of Disruptive Traffic Events

nitc Collaborator: Ran Wei 2017-2019

This project is funded by National Institute of Transportation and Communities.

Current traffic management strategies are based on expected conditions caused by recurring congestion (e.g. by time of day, day of week), and can be very effective when provisions are also given for reasonable variations from such expectations. However, traffic variations due to non-recurrent events (e.g. crashes) can be much larger and difficult to predict, making also challenging efforts to measure and forecast their disruptive effects. This project explores a proactive approach to manage non-recurring congestion by quantifying and modeling the effects of disruptive traffic events (crashes, major sporting events, weather, etc.) at a microscopic level using a comprehensive set of data sources. Outcomes from this research will lead to detailed event-based spatio-temporal congestion and safety models, ultimately enabling informed and proactive traffic management and safety countermeasures.

Spatial Optimization

Strategic planning and design for electric bus systems

USDOT uta PI: Ran Wei 2016-2018

This project is funded by U.S. Department of Transportation and Utah Transit Authorities.

Electric bus with zero-emission has been recognized as a promising alternative to diesel and compressed natural gas (CNG) bus to advance air quality and save fuel costs. The adoption of electric buses requires significant investment and needs strategic and comprehensive planning on how to deploy electric buses and associated infrastructure (e.g., charging stations). Important decisions in deploying electric buses and charging stations will include, among others, identifying appropriate driving range (battery specification) for electric buses, allocating electric buses to appropriate transit routes, and determining locations of charging stations and their corresponding capacities that can charge the electric buses in a cost and time-effective way. This research will use and develop a combination of GIS and optimization methods to help transit agencies make informed decisions regarding strategic planning and design for electric bus systems.

Evaluating and Enhancing Public Transit Systems for Operational Efficiency, Service Quality and Access Equity

nitc PI: Ran Wei 2016-2018

This project is funded by National Institute of Transportation and Communities.

Regular assessment and improvement of the performance of public transit systems are essential for transit service providers given limited funding and growing public needs. However, the evaluation of transit system performance is complex and challenging due to the diverse and competing goals of a public transportation agency, such as improving operational efficiency, increasing service quality, and providing equitable and just transit services. While previous work has examined public transit systems for these goals separately, the interplay of all three has not been investigated in research or practice. The proposed project will develop a comprehensive framework and an open-source toolbox for evaluating and enhancing the overall performance of public transit systems by using a combination of mathematical programming methods, GIS-based analysis and multi-objective spatial optimization techniques.

Enhancing Community Resilience and Optimizing Oil Spill Response through the Participatory Design of a Decision Support System

NASEM_logo PI: Tony Grubesic & Ran Wei 2016-2018

This project is funded by The National Academies of Sciences, Engineering, and Medicine .

In the United States, there is a large and growing interest in the vulnerability of complex social, economic and environmental systems. For example, consider the impacts of the BP Deepwater Horizon oil spill of 2010. In addition to the release of ~4.5 million barrels of oil into the Gulf of Mexico (GOM), eleven workers on the platform died, the spill lasted nearly three months, and at its peak, almost 87,000 square miles of fishable waters were closed. With 537,000 ocean sector employees and nearly $98 billion in gross domestic product (GDP) tied to these waters, the impacts of the Deepwater Horizon explosion and oil spill on the U.S. economy, environment, and hundreds of communities ringing the GOM was significant. Questions pertaining to the resilience of coastal communities, broadly defined (e.g., social, economic, environmental, cultural), are important when developing mitigation strategies for extreme events. So too are the operational challenges associated with allocating and dispatching human resources, equipment and supplies to areas impacted by a disaster. The purpose of this project is to develop a new, open source, spatial decision support system (SDSS) that will minimize the environmental, economic and social impacts of oil spills by optimizing the allocation of response crews and equipment, in both the marine and terrestrial environment.

Big Spatial Data

RIDIR: Scalable Geospatial Analytics for Social Science Research

nsf PI: Sergio Rey & Co-PI: Ran Wei, Amr Magdy, Vassilis Tsotras Oct 1, 2018 - Sept 30, 2021

The project's goals are to enhance the open source spatial analysis library PySAL in four main directions. The first innovation concerns new methods for the treatment of uncertainty in the American Community Survey data. Second, a new framework for statistical inference in the space-time analysis of segregation will be developed. Third, scalable algorithms for regionalization and spatially explicit aggregation of areal units will be produced. Finally, the project will develop a new scalable spatio-temporal data management layer to support the application of these three sets of analytics to large datasets. The software developed here will be included in enhanced (and new) modules in PySAL. This will afford the development of new types of applications using different interfaces from plug-ins for desktop GIS, both open source and proprietary, to the development of software as service systems and the ability of installations where on-premise requirements are binding (Census Research Data Centers). This portfolio of use cases will support a wide range of scientific disciplines and will provide direct benefits to society through new geospatial tools that will improve the state of the art in urban planning, economic development, and spatial public policy.

STAT: Social-Transportation Analytic Toolbox for Enhancing Multimodal Transportation Network

nitc PI: Ran Wei 2017-2019

This project is funded by National Institute of Transportation and Communities.

This research project will build an open-source socio-transportation analytic (STAT) toolbox for public transit system planning, in an effort to integrate social media and general transit feed specification (GTFS) data for transit agencies in evaluating and enhancing the performance of public transit system. This toolbox is novel and essential to transit agencies in two aspects. First, it enables the integration, analysis and visualization of two major new open transportation data: social media and GTFS data, to support transit decision making. Second, it allows transit agencies to evaluate service network efficiency and access equity of transit systems in a cohesive manner, and identify areas for improvement to better achieve these multi-dimensional objectives. The toolbox will employ a combination of data mining, geographical information systems, and transportation network modeling.

Network Effects of Disruptive Traffic Events

nitc Collaborator: Ran Wei 2017-2019

This project is funded by National Institute of Transportation and Communities.

Current traffic management strategies are based on expected conditions caused by recurring congestion (e.g. by time of day, day of week), and can be very effective when provisions are also given for reasonable variations from such expectations. However, traffic variations due to non-recurrent events (e.g. crashes) can be much larger and difficult to predict, making also challenging efforts to measure and forecast their disruptive effects. This project explores a proactive approach to manage non-recurring congestion by quantifying and modeling the effects of disruptive traffic events (crashes, major sporting events, weather, etc.) at a microscopic level using a comprehensive set of data sources. Outcomes from this research will lead to detailed event-based spatio-temporal congestion and safety models, ultimately enabling informed and proactive traffic management and safety countermeasures.

Spatial Uncertainty

RIDIR: Scalable Geospatial Analytics for Social Science Research

nsf PI: Sergio Rey & Co-PI: Ran Wei, Amr Magdy, Vassilis Tsotras Oct 1, 2018 - Sept 30, 2021

The project's goals are to enhance the open source spatial analysis library PySAL in four main directions. The first innovation concerns new methods for the treatment of uncertainty in the American Community Survey data. Second, a new framework for statistical inference in the space-time analysis of segregation will be developed. Third, scalable algorithms for regionalization and spatially explicit aggregation of areal units will be produced. Finally, the project will develop a new scalable spatio-temporal data management layer to support the application of these three sets of analytics to large datasets. The software developed here will be included in enhanced (and new) modules in PySAL. This will afford the development of new types of applications using different interfaces from plug-ins for desktop GIS, both open source and proprietary, to the development of software as service systems and the ability of installations where on-premise requirements are binding (Census Research Data Centers). This portfolio of use cases will support a wide range of scientific disciplines and will provide direct benefits to society through new geospatial tools that will improve the state of the art in urban planning, economic development, and spatial public policy.

Spatial Pattern Discovery for Uncertain Data from the American Community Survey

utah PI: Ran Wei 2015-2017

This project is funded by University of Utah.

Starting from 2010, the American Community Survey (ACS) completely replaced the decennial census long form and became the primary data source for detailed characteristics of the U.S. population. The annual ACS estimates also present great challenges to data sampling and inferences to be made based on the data. Given the limited budget and time constraints, the ACS utilizes a much smaller sample (1 in 38 households) than the decennial long form (1 in 6 households), resulting in a potentially much larger sampling error. A growing number of literature have recognized the issue and discussed the uncertainty involved, highlighting the necessity for ACS data users to understand the data quality and its implications in future analysis and inferences, lest the planning and decision-making be biased and unreliable. While data error or uncertainty in ACS estimates has been widely acknowledged, little has been done to devise methods accounting for such error or uncertainty. This project will develop a suite of advanced mathematical, spatial statistical and geovisualization methods that explicitly take into account data uncertainty in ACS estimates to accurately and precisely identify significant patterns of demographic and socioeconomic characteristics across space.

Open Source & Science

RIDIR: Scalable Geospatial Analytics for Social Science Research

nsf PI: Sergio Rey & Co-PI: Ran Wei, Amr Magdy, Vassilis Tsotras Oct 1, 2018 - Sept 30, 2021

The project's goals are to enhance the open source spatial analysis library PySAL in four main directions. The first innovation concerns new methods for the treatment of uncertainty in the American Community Survey data. Second, a new framework for statistical inference in the space-time analysis of segregation will be developed. Third, scalable algorithms for regionalization and spatially explicit aggregation of areal units will be produced. Finally, the project will develop a new scalable spatio-temporal data management layer to support the application of these three sets of analytics to large datasets. The software developed here will be included in enhanced (and new) modules in PySAL. This will afford the development of new types of applications using different interfaces from plug-ins for desktop GIS, both open source and proprietary, to the development of software as service systems and the ability of installations where on-premise requirements are binding (Census Research Data Centers). This portfolio of use cases will support a wide range of scientific disciplines and will provide direct benefits to society through new geospatial tools that will improve the state of the art in urban planning, economic development, and spatial public policy.

Comparative Regional Inequality Dynamics: Multiscalar and Multinational Perspectives

nsf PI: Sergio Rey & Co-PI: Yehua Wei May 1, 2018 - Apr 30, 2021

This research project will comparatively analyze the dynamics of regional inequality in the United States and China using recent developments in geographic information science. Because the investigators will focus on core-periphery structure, place mobility, and scalar effects, this project will significantly improve basic understanding of spatial inequality. This project will advance knowledge about the dynamics of regional income inequality in a number of different ways. It will provide a systematic study of the multi-scalar nature of regional inequality, thereby providing a clearer picture of the dynamics of regional inequality and its relationship to scale. It will provide new insights about place mobility and about the core-periphery structure in regional development, and it will provide new perspectives regarding the role of geography in regional inequality. Project findings should prove valuable to domestic and international organizations, such as non-profit organizations and the U.S. and Chinese governments, giving them improved understanding of inequality and facilitating the development of more informed policies that address development issues in equitable ways.

Neighborhoods in Space-Time Contexts

nsf PI: Sergio Rey Sept 1, 2017 - Aug 31, 2020 Learn more

This research project will advance the way in which neighborhoods are defined in urban social science research. The definition of a socioeconomic neighborhood is an important and central issue across many research domains. Neighborhoods serve as the organizational units to frame empirical research that has examined a wide array of issues, including the ability of social networks to produce collective efficacy in a spatial context, the relationships between concentrated poverty traps and violence, spatial sorting and segregation, and the role of neighborhoods as seeds for wider urban economic development. Despite this importance, consensus has not yet emerged regarding how to operationalize neighborhoods in practice. Moreover, the role of spatial structure largely has been ignored in existing approaches to neighborhood definition. This project will produce new methods for neighborhood identification and analysis that draw on recent developments in geographic information science and spatial statistics. These new methods will enhance the urban social science toolkit enabling researchers to carry out empirical work that can be replicated, thereby moving urban policy research onto stronger analytic foundations. The new methods and analytics will be implemented in a publicly available open-source longitudinal neighborhood analysis package. The project will provide training to a post-doctoral research associate and will include a partnership between academia and industry, with applications to commercial market segmentation and analysis.

New Approaches to Spatial Distribution Dynamics

nsf PI: Sergio Rey Jul 31, 2014 - Aug 31, 2018 Learn more

This research project will improve understanding of spatial inequality dynamics through methodological advances in measurement and modeling. Understanding the nature of spatial inequality dynamics is vital to both basic social science and to public policy, yet existing methods and models provide incomplete views of these spatial dynamics. While a central focus of inequality research has been on the evolution of the aggregate income distribution, much less attention has been directed at the spatial pattern of inequality and pattern dynamics. Spatial inequalities can have important implications for social cohesion, economic growth, and the design of policies targeted at reducing the level of inequality. The advances produced in the project will have wide applicability. In addition to spatial income inequality dynamics, many other social and economic phenomena have distributions that evolve in space and time. Software packages will be delivered as open-source projects and accompanied with extensive tutorials and documentation to facilitate broad dissemination across the social sciences.

Evaluating and Enhancing Public Transit Systems for Operational Efficiency, Service Quality and Access Equity

nitc PI: Ran Wei 2016-2018

This project is funded by National Institute of Transportation and Communities.

Regular assessment and improvement of the performance of public transit systems are essential for transit service providers given limited funding and growing public needs. However, the evaluation of transit system performance is complex and challenging due to the diverse and competing goals of a public transportation agency, such as improving operational efficiency, increasing service quality, and providing equitable and just transit services. While previous work has examined public transit systems for these goals separately, the interplay of all three has not been investigated in research or practice. The proposed project will develop a comprehensive framework and an open-source toolbox for evaluating and enhancing the overall performance of public transit systems by using a combination of mathematical programming methods, GIS-based analysis and multi-objective spatial optimization techniques.

Enhancing Community Resilience and Optimizing Oil Spill Response through the Participatory Design of a Decision Support System

NASEM_logo PI: Tony Grubesic & Ran Wei 2016-2018

This project is funded by The National Academies of Sciences, Engineering, and Medicine .

In the United States, there is a large and growing interest in the vulnerability of complex social, economic and environmental systems. For example, consider the impacts of the BP Deepwater Horizon oil spill of 2010. In addition to the release of ~4.5 million barrels of oil into the Gulf of Mexico (GOM), eleven workers on the platform died, the spill lasted nearly three months, and at its peak, almost 87,000 square miles of fishable waters were closed. With 537,000 ocean sector employees and nearly $98 billion in gross domestic product (GDP) tied to these waters, the impacts of the Deepwater Horizon explosion and oil spill on the U.S. economy, environment, and hundreds of communities ringing the GOM was significant. Questions pertaining to the resilience of coastal communities, broadly defined (e.g., social, economic, environmental, cultural), are important when developing mitigation strategies for extreme events. So too are the operational challenges associated with allocating and dispatching human resources, equipment and supplies to areas impacted by a disaster. The purpose of this project is to develop a new, open source, spatial decision support system (SDSS) that will minimize the environmental, economic and social impacts of oil spills by optimizing the allocation of response crews and equipment, in both the marine and terrestrial environment.

STAT: Social-Transportation Analytic Toolbox for Enhancing Multimodal Transportation Network

nitc PI: Ran Wei 2017-2019

This project is funded by National Institute of Transportation and Communities.

This research project will build an open-source socio-transportation analytic (STAT) toolbox for public transit system planning, in an effort to integrate social media and general transit feed specification (GTFS) data for transit agencies in evaluating and enhancing the performance of public transit system. This toolbox is novel and essential to transit agencies in two aspects. First, it enables the integration, analysis and visualization of two major new open transportation data: social media and GTFS data, to support transit decision making. Second, it allows transit agencies to evaluate service network efficiency and access equity of transit systems in a cohesive manner, and identify areas for improvement to better achieve these multi-dimensional objectives. The toolbox will employ a combination of data mining, geographical information systems, and transportation network modeling.

ProjectsΒΆ