Simulation-based studies are frequently used for urban planning, traffic operation, and emergency... more Simulation-based studies are frequently used for urban planning, traffic operation, and emergency management. Given the urban transportation systems complexity and data availability, most traffic simulation models focus on selected geographic areas. With regularly improved Volunteered Geographic Information (VGI), such as OpenStreetMap, and annual updated LandScanTM global population distribution data, we present TUMS, a Toolbox for Urban Mobility Simulations at global scale. TUMS uses unified data structure for simulation inputs. For studies in United States, both daily traffic simulation and emergency evacuation simulation are modeled. The origin-destination matrixes for daily traffic are extracted from Census Transportation Planning Products (CTPP). Both private vehicles and public buses are included in urban transportation systems. A super-node trip distribution model is developed to simulate evacuation scenarios with good computational performance. International studies only implement evacuation scenarios at current stage due to the lack of open transportation data. Two levels of visualization tools are implemented for system analysis, including link-based macroscopic visualization and vehicle-based microscopic visualization. We modeled five American cities (Washington D.C., Alexandria in Virginia, Fort Belvior in Virginia, Knox county in Tennessee, and Seattle in Washington) and seven international cities (Beijing in China, Chengdu in China, Seoul in South Korea, Singapore, Taipei in Taiwan, Kharhoum in Sudan, Juba in South Sudan, Doha in Qatar, Berlin in Germany). TUMS provides a unified platform for global urban mobility researchers to estimate and visualize the performance of transportation systems under daily or emergency scenarios.
Mobility has been an important measure to evaluate the efficiency and livability in urban areas. ... more Mobility has been an important measure to evaluate the efficiency and livability in urban areas. With the increasing amount of volunteered geographic information, such as OpenStreetMap, we can build applications to study urban mobility and dynamics at national and even global level. By taking advantage of the LandScan high-resolution population distribution data (30m by 30m in United States and 1 km by 1 km in other counties), we developed the TUMS (Toolbox for Urban Mobility Simulations) system, a microsimulaiton approach to address transportation problems at global level. The TUMS integrates the National Household Travel Survey (NHTS) data for travel demand modeling in US case studies. To demonstrate the system capability and performance, two kinds of traffic scenarios in Alexandria, Virginia are modeled: daily normal traffic scenario and emergency evacuation scenario. The first scenario is modeled with selected NHTS data and simulated with microscopic traffic models in the TUMS. The second scenario is modeled based on the no-notice emergency evacuation assumption – everyone leaves the affected area with shortest paths. Through the comparison study and vehicle-based microscopic visualization, users can easily find the urban mobility patterns under different scenarios. This can help transportation planners to identify system bottlenecks and to test different operation strategies for special events, such as game-day traffic operation and disaster management. The TUMS system is easily scaled to any geographic locations with its unique GIS data resources. With proper analysis and process of NHTS data, it is fairly fast to simulation traffic mobility in a multiple-county megaregion area.
Natural or man-made disasters (e.g. Atlanta ice snow in 2014, Boston Marathon bombings in 2013) h... more Natural or man-made disasters (e.g. Atlanta ice snow in 2014, Boston Marathon bombings in 2013) have tremendous impacts on urban transportation systems. How to maintain essential transportation operations during critical infrastructure disruption is an important issue. Due to the tremendous cost of large scale evacuation practice and lack of sufficient real-world evacuation data, simulation-based studies can help emergency managers and transportation planners to estimate the evacuation clearance time and potential cost in a fast-than-real-time manner. In order to optimize both performance accuracy and computing speed, we proposed an efficient framework World-Wide Emergency Evacuation (WWEE) to simulate transportation systems in global urban areas under emergency evacuation.
Transportation simulations are frequently used for mobility assessment urban planning, traffic op... more Transportation simulations are frequently used for mobility assessment urban planning, traffic operations, and emergency management. Given urban transportation systems complexity and data availability, most traffic simulation models focus on a selected geographic area. With increasing availability and quality of traditional and crowdsourced data, we have utilized the OpenStreetMap roads network, and have integrated high resolution population LandScanTM data with traffic simulation to create a Toolbox for Urban Mobility Simulations (TUMS) at global scale. TUMS consists of three major components: data processing, traffic simulation models, and Internet-based visualizations. It integrates OpenStreetMap, LandScanTM population, and other open data (Census Transportation Planning Products, National Household Travel Survey, etc.) to generate both normal traffic operation and emergency evacuation scenarios. TUMS integrates TRANSIMS and MITSIM as traffic simulation engines, which are open-source and widely-accepted for scalable traffic simulations. It provides megacity level large-scale microscopic traffic simulation, flexibility for rapid local adoptions, open-source based platform, and extensibility to multiple applications. Besides private vehicles, TUMS has the capability to model public buses in urban transportation systems. Two levels of visualization tools are implemented for system analysis, including link-based macroscopic visualization and vehicle-based microscopic visualization. We modeled six US cities (Washington D.C., Alexandria, VA, Fort Belvior, VA, Nashville, TN, Knox County, TN, and Seattle, WA) and eight international cities (Beijing, China, Chengdu, China, Seoul, South Korea, Singapore, Taipei, Taiwan, Kharhoum, Sudan, Juba, South Sudan, Doha, Qatar, Berlin, Germany, and Dakar, Senegal). We also implemented a real-time geo-tagged tweets visualization tool to display the locations of transportation related tweets. This will help us to identify potential transportation network problems in urban areas. This platform uses node.js and Twitter API as the basic technologies for implementation. Some preliminary results in Los Angeles are displayed. Further work on integration this data into the TUMS system will improve system performance. Together with geo-tagged tweets data, TUMS can provide unified platform for global urban mobility researchers to estimate and visualize the performance of transportation systems under daily or emergency scenarios.
Simulation-based studies are frequently used for evacuation planning and decision making processe... more Simulation-based studies are frequently used for evacuation planning and decision making processes. Given the transportation systems complexity and data availability, most evacuation simulation models focus on certain geographic areas. With routine improvement of OpenStreetMap road networks and LandScanTM global population distribution data, we present WWEE, a uniform system for world-wide emergency evacuation simulations. WWEE uses unified data structure for simulation inputs. It also integrates a super-node trip distribution model as the default simulation parameter to improve the system computational performance. Two levels of visualization tools are implemented for evacuation performance analysis, including link-based macroscopic visualization and vehicle-based microscopic visualization. For left-hand and right-hand traffic patterns in different countries, the authors propose a mirror technique to experiment with both scenarios without significantly changing traffic simulation models. Ten cities in US, Europe, Middle East, and Asia are modeled for demonstration. With default traffic simulation models for fast and easy-to-use evacuation estimation and visualization, WWEE also retains the capability of interactive operation for users to adopt customized traffic simulation models. For the first time, WWEE provides a unified platform for global evacuation researchers to estimate and visualize the performance of transportation systems under evacuation scenarios.
Explosion of spatial data from satellite to citizen sensors has posed the critical challenge of B... more Explosion of spatial data from satellite to citizen sensors has posed the critical challenge of Big Spatial Data integration, analysis, and visualization. This article focuses on research and development activities at Oak Ridge National Laboratory (ORNL) that are addressing end-user applications utilizing high performance computing based geospatial science and technology solutions to optimize the analysis, modeling, and multi-megapixel scale visualization of the geospatial data. Specifically we highlight recent developments and successes in the areas of high resolution settlement mapping, transportation and mobility analysis, and effective monitoring of biomass for energy and food security.
Simulation-based studies are frequently used for urban planning, traffic operation, and emergency... more Simulation-based studies are frequently used for urban planning, traffic operation, and emergency management. Given the urban transportation systems complexity and data availability, most traffic simulation models focus on selected geographic areas. With regularly improved Volunteered Geographic Information (VGI), such as OpenStreetMap, and annual updated LandScanTM global population distribution data, we present TUMS, a Toolbox for Urban Mobility Simulations at global scale. TUMS uses unified data structure for simulation inputs. For studies in United States, both daily traffic simulation and emergency evacuation simulation are modeled. The origin-destination matrixes for daily traffic are extracted from Census Transportation Planning Products (CTPP). Both private vehicles and public buses are included in urban transportation systems. A super-node trip distribution model is developed to simulate evacuation scenarios with good computational performance. International studies only implement evacuation scenarios at current stage due to the lack of open transportation data. Two levels of visualization tools are implemented for system analysis, including link-based macroscopic visualization and vehicle-based microscopic visualization. We modeled five American cities (Washington D.C., Alexandria in Virginia, Fort Belvior in Virginia, Knox county in Tennessee, and Seattle in Washington) and seven international cities (Beijing in China, Chengdu in China, Seoul in South Korea, Singapore, Taipei in Taiwan, Kharhoum in Sudan, Juba in South Sudan, Doha in Qatar, Berlin in Germany). TUMS provides a unified platform for global urban mobility researchers to estimate and visualize the performance of transportation systems under daily or emergency scenarios.
Mobility has been an important measure to evaluate the efficiency and livability in urban areas. ... more Mobility has been an important measure to evaluate the efficiency and livability in urban areas. With the increasing amount of volunteered geographic information, such as OpenStreetMap, we can build applications to study urban mobility and dynamics at national and even global level. By taking advantage of the LandScan high-resolution population distribution data (30m by 30m in United States and 1 km by 1 km in other counties), we developed the TUMS (Toolbox for Urban Mobility Simulations) system, a microsimulaiton approach to address transportation problems at global level. The TUMS integrates the National Household Travel Survey (NHTS) data for travel demand modeling in US case studies. To demonstrate the system capability and performance, two kinds of traffic scenarios in Alexandria, Virginia are modeled: daily normal traffic scenario and emergency evacuation scenario. The first scenario is modeled with selected NHTS data and simulated with microscopic traffic models in the TUMS. The second scenario is modeled based on the no-notice emergency evacuation assumption – everyone leaves the affected area with shortest paths. Through the comparison study and vehicle-based microscopic visualization, users can easily find the urban mobility patterns under different scenarios. This can help transportation planners to identify system bottlenecks and to test different operation strategies for special events, such as game-day traffic operation and disaster management. The TUMS system is easily scaled to any geographic locations with its unique GIS data resources. With proper analysis and process of NHTS data, it is fairly fast to simulation traffic mobility in a multiple-county megaregion area.
Natural or man-made disasters (e.g. Atlanta ice snow in 2014, Boston Marathon bombings in 2013) h... more Natural or man-made disasters (e.g. Atlanta ice snow in 2014, Boston Marathon bombings in 2013) have tremendous impacts on urban transportation systems. How to maintain essential transportation operations during critical infrastructure disruption is an important issue. Due to the tremendous cost of large scale evacuation practice and lack of sufficient real-world evacuation data, simulation-based studies can help emergency managers and transportation planners to estimate the evacuation clearance time and potential cost in a fast-than-real-time manner. In order to optimize both performance accuracy and computing speed, we proposed an efficient framework World-Wide Emergency Evacuation (WWEE) to simulate transportation systems in global urban areas under emergency evacuation.
Transportation simulations are frequently used for mobility assessment urban planning, traffic op... more Transportation simulations are frequently used for mobility assessment urban planning, traffic operations, and emergency management. Given urban transportation systems complexity and data availability, most traffic simulation models focus on a selected geographic area. With increasing availability and quality of traditional and crowdsourced data, we have utilized the OpenStreetMap roads network, and have integrated high resolution population LandScanTM data with traffic simulation to create a Toolbox for Urban Mobility Simulations (TUMS) at global scale. TUMS consists of three major components: data processing, traffic simulation models, and Internet-based visualizations. It integrates OpenStreetMap, LandScanTM population, and other open data (Census Transportation Planning Products, National Household Travel Survey, etc.) to generate both normal traffic operation and emergency evacuation scenarios. TUMS integrates TRANSIMS and MITSIM as traffic simulation engines, which are open-source and widely-accepted for scalable traffic simulations. It provides megacity level large-scale microscopic traffic simulation, flexibility for rapid local adoptions, open-source based platform, and extensibility to multiple applications. Besides private vehicles, TUMS has the capability to model public buses in urban transportation systems. Two levels of visualization tools are implemented for system analysis, including link-based macroscopic visualization and vehicle-based microscopic visualization. We modeled six US cities (Washington D.C., Alexandria, VA, Fort Belvior, VA, Nashville, TN, Knox County, TN, and Seattle, WA) and eight international cities (Beijing, China, Chengdu, China, Seoul, South Korea, Singapore, Taipei, Taiwan, Kharhoum, Sudan, Juba, South Sudan, Doha, Qatar, Berlin, Germany, and Dakar, Senegal). We also implemented a real-time geo-tagged tweets visualization tool to display the locations of transportation related tweets. This will help us to identify potential transportation network problems in urban areas. This platform uses node.js and Twitter API as the basic technologies for implementation. Some preliminary results in Los Angeles are displayed. Further work on integration this data into the TUMS system will improve system performance. Together with geo-tagged tweets data, TUMS can provide unified platform for global urban mobility researchers to estimate and visualize the performance of transportation systems under daily or emergency scenarios.
Simulation-based studies are frequently used for evacuation planning and decision making processe... more Simulation-based studies are frequently used for evacuation planning and decision making processes. Given the transportation systems complexity and data availability, most evacuation simulation models focus on certain geographic areas. With routine improvement of OpenStreetMap road networks and LandScanTM global population distribution data, we present WWEE, a uniform system for world-wide emergency evacuation simulations. WWEE uses unified data structure for simulation inputs. It also integrates a super-node trip distribution model as the default simulation parameter to improve the system computational performance. Two levels of visualization tools are implemented for evacuation performance analysis, including link-based macroscopic visualization and vehicle-based microscopic visualization. For left-hand and right-hand traffic patterns in different countries, the authors propose a mirror technique to experiment with both scenarios without significantly changing traffic simulation models. Ten cities in US, Europe, Middle East, and Asia are modeled for demonstration. With default traffic simulation models for fast and easy-to-use evacuation estimation and visualization, WWEE also retains the capability of interactive operation for users to adopt customized traffic simulation models. For the first time, WWEE provides a unified platform for global evacuation researchers to estimate and visualize the performance of transportation systems under evacuation scenarios.
Explosion of spatial data from satellite to citizen sensors has posed the critical challenge of B... more Explosion of spatial data from satellite to citizen sensors has posed the critical challenge of Big Spatial Data integration, analysis, and visualization. This article focuses on research and development activities at Oak Ridge National Laboratory (ORNL) that are addressing end-user applications utilizing high performance computing based geospatial science and technology solutions to optimize the analysis, modeling, and multi-megapixel scale visualization of the geospatial data. Specifically we highlight recent developments and successes in the areas of high resolution settlement mapping, transportation and mobility analysis, and effective monitoring of biomass for energy and food security.
Global urbanization has boosted the population growth unprecedentedly, which has also increased t... more Global urbanization has boosted the population growth unprecedentedly, which has also increased the travel demand and transportation infrastructure demand in urban areas. Urban mobility is one of the important measures to evaluate the efficiency and livability in certain geographic areas. The American Community Survey (ACS) 5-year data and the Census Transportation Planning Products (CTPP) special tabulation provide home-to-work traffic flow at Traffic Analysis Zones (TAZ) level. This provides necessary inputs for link-based mobility analysis with microscopic traffic simulations. In order to provide an open-source and easy-to-use tool to urban dynamics and transportation research communities, we developed the TUMS (Toolbox for Urban Mobility Simulation) system, which is a unique high-resolution population distribution and traffic microsimulation based approach. There are three major components in TUMS: data processing module, traffic modeling and simulation module, and web-based visualization module. The major input data include LandScan USA high resolution population distribution dataset, OpenStreetMap road network, TAZ polygons, and CTPP traffic flow at TAZ level. We use CTPP traffic flow data as our major input for daily traffic demand modeling. The high margins of error in TAZ based traffic flow questioned the modeling and simulation accuracy. In the case study of Knoxville, Tennessee traffic modeling process, we use the mean values from CTPP to build origin-destination tables. We run the whole microscopic traffic simulation and summarize the road network performance at link level. To compare the simulation results, we also run the simulation with LandScan USA high resolution dataset. The LandScan USA has population distribution in both daytime and nighttime. The resolution is 30 meters by 30 meters. Each cell has the number of population. We built the traffic flow using the difference between nighttime population and daytime population. We aggregated the LandScan cells to TAZ to build the TAZ based origin-destination table again. Through these two simulations, we find there is no significant difference on the overall road network performance. The difference varies at link level. For the very first time, the CTPP data provide an approach for microscopic traffic simulation, which can improve the model accuracy and details. One way to improve the CTPP traffic flow accuracy is randomly sampling the margins of error based on the weights of aggregated LandScan USA dataset.
The emerging discipline of Computational Transportation Science integrates computer science, comp... more The emerging discipline of Computational Transportation Science integrates computer science, computational and data science with transportation modeling, planning, and economic aspects to leverage developments in the above domains. By taking advantage of Big Data analysis and high performance computing, Computational Transportation Science applications have a unique opportunity to build more efficient and sustainable transportation systems and communities. At Oak Ridge National Laboratory (ORNL), researchers are attempting to address urban mobility issues at global scale by utilizing ORNL leading computing and data resources. We will describe the development of the TUMS (Toolbox for Urban Mobility Simulations) system, a unique high-resolution population distribution and traffic microsimulation based approach, and its application in emergency evacuation scenarios. We will discuss the challenges and opportunities in the traffic modeling and simulation field from the experiences of the TUMS development and also share insights and perspectives on the recent trends in computational transportation science.
Uploads