The National Institute for Transportation and Communities (NITC) is coming to a close, and we wanted to take a moment to look back at some of the impacts our center has had. See below for some of the outcomes of NITC research improving transportation data and tools, to help practitioners create research-driven solutions.

• First, read about VisionEval, the open-source modeling framework which has been enhanced through work supported by NITC.
• Next, check out an overview of BikePed Portal, the nationwide nonmotorized database which was launched by a NITC pooled fund project.
• Finally, learn how another NITC pooled fund project gave rise to a "data fusion" technique which is making it possible for agencies to predict bicycle volumes across an entire transportation network.

Strengthening the Predictive Powers of VisionEval: NITC's Impact On Scenario Planning

VisionEval is an open-source modeling framework used by transportation agencies all around the country to evaluate the long-term impacts of various transportation, land use, and policy scenarios. Over the past several years, the US Department of Transportation-funded National Institute for Transportation and Communities (NITC) has helped to refine VisionEval, incorporating new travel modes such as car sharing, bike sharing, ride hailing, and autonomous vehicles, as well as active transportation activity.

"My PhD advisor used to say that VisionEval is a SIM City for professionals," said Liming Wang, referring to the computer game in which players build and manage a virtual city. By anticipating the effects of different policies and investments, it serves as a decision support system for planners, policymakers and the general public. Wang has led several NITC research efforts to refine and enhance VisionEval.

HOW HAS NITC IMPROVED VISIONEVAL?

Expanding the power of VisionEval offers a multitude of benefits for state and regional planning. In addition to Liming Wang, NITC researchers Kelly Clifton, Jennifer Dill, Jenny Liu, Alex Bigazzi, Yao-Jan Wu, Joseph Broach and Kristina Currans have all worked to grow VisionEval's capabilities in various ways.

Incorporating Bicycle Activity and Vehicle Travel Reduction from Bicycle Infrastructure

Principal Investigators: Joe Broach, Portland State University; Kristina Currans, University of Arizona

In existing versions of VisionEval, increases in cycling and reductions in vehicle miles traveled (VMT) were not sensitive to  improvements in bicycle network connections. This project was significant because the analysis provided a framework that linked such infrastructure changes to bike travel, accessibility, and VMT outcomes in a way that could also be incorporated into the VisionEval planning tool. This could significantly improve the representation of bicycle travel behavior for scenario planning purposes.

Land Use and Transportation Policies for a Sustainable Future with Autonomous Vehicles 

Principal Investigators: Liming Wang, Portland State University; Yao-Jan Wu, University of Arizona

Even though there are tremendous uncertainties in the timing and evolution path of the Autonomous Vehicles (AV) technology, it may become a likely reality within most MPOs' long-range regional transportation plan horizon of twenty years. This project studied the possible impacts on travel and land use of emerging AV technology and focuses on advancing this innovative mobility option by making sure it serves the greater good of building sustainable and equitable communities in its adoption. The project also contributed to smart cities research by examining the likely social, economic, and environmental outcomes of integrating AV in our cities and by starting to consider policies and plans to preempt their potential adverse impacts and to ensure that the AV technology improves access for all people.

Incorporate Emerging Travel Modes in the Regional Strategic Planning Model (RSPM) Tool

Principal Investigators: Liming Wang and Jennifer Dill, Portland State University; Kelly Clifton, Portland State University (Now at University of British Columbia)

This project offers help to planners seeking to incorporate emerging travel modes—including car sharing, bike sharing, ride hailing, and autonomous vehicles—into regional travel demand models. The research team completed a nationwide survey and collected data from 1,117 valid participants. The data were examined to learn about their recent travel behaviors, attitude, and their stated preferences about using emerging travel modes when presented along with their chosen mode. The team then developed models using the data from the stated preference choice experiments, along with information from the revealed preferences and socio-demographic characteristics of the survey respondents. 

Informed by their earlier NITC-funded research, Wang and Jenny Liu of PSU also worked with the Oregon Department of Transportation (ODOT) to develop a Transportation Cost Index: A Comprehensive Performance Measure for Transportation and Land Use Systems. The index they developed helps DOTs and MPOs to adopt comprehensive measures in the decision process which allow consistent comparison of multimodal performance over time and across geographic areas. It is applicable for monitoring historical and projected trends (benchmarking), evaluating and comparing outcomes from what-if scenarios (scenario evaluation), as well as reporting on current status.

In another ODOT project currently underway, Wang is now working to incorporate that cost index into VisionEval. The research goal is to upgrade the land use model in the current VisionEval implementation and expand its capacity to enable evaluation of transportation and land use policies. The upgrades will improve ODOT’s abilities to evaluate how land use and transportation strategies will affect each other and GHG emissions. In addition, they will greatly improve capabilities for modeling land use and its consequences at a more detailed level, evaluating regional land use strategies, and evaluating the social equity implications of land use and transportation strategies.. Read about that project: Development Of A New VisionEval Land Use Model And Applications To Evaluation Of Climate Strategies In Oregon.

HISTORY AND BACKGROUND

Originally developed by Brian Gregor at the Oregon Department of Transportation (ODOT) in the late 1990s, the open-source code of VisionEval has since been widely adopted and adapted for analysis of multiple factors such as energy consumption, emissions, and travel demand. Its modular design enables users to tailor the model to specific regional or project needs, helping planners make data-driven decisions to shape sustainable transportation systems.

By analyzing energy use, emissions, and fuel consumption under different scenarios, VisionEval helps planners develop strategies to reduce greenhouse gas emissions and meet sustainability targets. It allows transportation professionals to evaluate how policies will affect different communities, particularly underserved populations, by factoring in equity considerations like accessibility and affordability. Since it is an open-source tool, agencies can adopt and modify it as needed for a low cost.

Read more about VisionEval's impacts in a September 2024 summary from Resource Systems Group (RSG), Strengthening Scenario Planning: How VisionEval Helps Planners Quantify the Measures that Matter. Another September 2024 article, this one from PSU's College of Urban and Public Affairs, explores VisionEval's versatility as a strategic planning tool.

Bike-Ped Portal: Making the Most of Nonmotorized Data

In recent years, cities and counties all across the United States have invested in bicycle and pedestrian counters. Having reliable data on non-motorized traffic can help jurisdictions make informed decisions about infrastructure investments, measure the effectiveness of biking and walking initiatives, and improve safety. But having a bike-ped count program is just the beginning. Once counts have been collected, cities face new challenges. How can multiple agencies share data among themselves? How can data be standardized and compared? And what is the most effective way to communicate what the numbers mean? Enter Bike-Ped Portal, the comprehensive non-motorized data management system for manual and automated non-vehicular multi-modal counts housed at Portland State University (PSU).

BikePed Portal was established in 2015 by Transportation Research and Education Center (TREC) researchers at Portland State University through a pooled fund grant administered by the National Institute for Transportation and Communities (NITC).This specialized data lake and management system makes sharing data—within an agency, with partners at other agencies, and with the public—easily accessible. Since its NITC-funded inception, it has grown into a unique national resource for bike and pedestrian data. In recent years, BikePed Portal has gained a dashboard for the national capital region and one for the Portland-Vancouver region, allowing users to see all the data for those regions in one place. 

Agencies who use Bike-Ped Portal may have counters from different vendors, or use more than one platform to manage data. That's why Bike-Ped Portal is able to accept data in multiple formats, whether that's a spreadsheet with manual counts, output from automated counters, or APIs such as EcoCounter. Once the system receives new data, it is converted into a standardized format. Quality control tools are available, along with the ability to add correction factors and data annotations.

Bike-Ped Portal does more than reduce the time and effort required to manage multiple data sources, format, standardize, validate, and share data openly. TREC's transportation data team can also build custom analytical tools and dashboards to help data tell a story. Transportation Data Program Manager Tammy Lee, and Senior Research Associate Basem Elazzabi, who head up TREC's transportation data program, work with agencies to build tools which meet their specific needs, help reduce redundancies and standardize reporting.

Data Fusion: Predicting Bicycle Volumes Using Machine Learning

In a series of research projects funded by the National Institute for Transportation and Communities (NITC), researchers have been developing new approaches that combine conventional and emerging data sources to estimate bicycle volumes. Having accurate bicycle volumes can help state departments of transportation (DOTs) and other agencies to prioritize projects, plan and design new bicycle infrastructure, and conduct safety analyses.

Traditional permanent and short-term counting methods have a high degree of accuracy but are limited to certain locations or short periods of time, while crowdsourced data (such as Strava or StreetLight) can cover a wider area but with less accuracy. Fusing the two methods together–potentially with the use of deep learning algorithms–is a promising way to get the best of both.

The latest report to come out of these efforts, by Sirisha Kothuri, Banafsheh Rekabdar and Joe Broach of Portland State University, pushed the needle forward on using advanced techniques to extrapolate data over a large transportation network. Two PSU graduate students also worked on the project: Saba Izadkhah, who is working toward a PhD in computer science, and Andrew Wagner, a computer science masters student.

"These methods are still evolving, and it's still in the research phase. But I think the time is not far off when we will start using these methods as more mainstream," Kothuri said.

The techniques have been continually developed and refined since this work first began in 2018. At that time, NITC launched a pooled fund project with support from the DOTs of Oregon, Virginia, Colorado, Utah, and the District of Columbia, as well as Central Lane MPO and the Portland Bureau of Transportation. With matching funds from NITC, those agencies came together to fund the initial project Exploring Data Fusion Techniques to Estimate Network-Wide Bicycle Volumes, with a research team led by Kothuri made up of researchers from PSU and the University of Texas at Arlington.

The new NITC report published this month, Improving the Accuracy and Precision of Bicycle Volume Estimates Using Advanced Machine Learning Approaches (PDF), shows that the latest models offer increased accuracy. A paper based on this work was presented at the Institute of Electrical and Electronics Engineers' International Conference on Artificial Intelligence x Science, Engineering and Technology at the beginning of October. 

"We know that for pedestrians, injuries and fatalities are at an all time high. Bicyclist safety is also of top concern. So these estimates are really critical for agencies right now," Kothuri said.

HOW DOES IT WORK?

The researchers train the model on existing count data from certain locations, then use that trained model to predict volumes at locations where there is count data that the model hasn't seen. They then compare the model's predictions with the actual count data to see how accurate it is. 

Using long short-term memory networks and deep neural networks, the method involves the combining of static variables—such as network characteristics, demographics, and land use— with dynamic crowdsourced data and count data from different regions. The research has shown that crowd-sourced data alone cannot replace traditional count data. For this method to work, you need both.

Regional data is key to the success of the model: the more local count data you have, and the more you can train the model to predict based on the local data, the better the accuracy will be. 

The model also fared better when using Monthly Average Daily Bicyclists (MADB) as a target, rather than Annual Average Daily Bicyclists (AADB), because breaking each counter down into monthly units gave it more data points.

"Basically, the more data it has, the smarter it gets," said Rekabdar. 

ABOUT THE PROJECT

For more details, watch a 2022 research highlight video about the original pooled fund study, or a recorded 2022 Webinar: Exploring Data Fusion Techniques to Derive Bicycle Volumes on a Network.

The National Institute for Transportation and Communities (NITC) is one of seven U.S. Department of Transportation national university transportation centers. NITC is a program of the Transportation Research and Education Center (TREC) at Portland State University. This PSU-led research partnership also includes the Oregon Institute of Technology, University of Arizona, University of Oregon, University of Texas at Arlington and University of Utah. We pursue our theme — improving mobility of people and goods to build strong communities — through research, education and technology transfer.