Lead presenter: Kyle Titlow, Bureau of Transportation Statistics
Biography:
Kyle Titlow is a postgraduate geospatial analyst fellow in the Office of Spatial Analysis and Visualization at the Bureau of Transportation Statistics (BTS) at the United States Department of Transportation in Washington, DC. Supported by the Oak Ridge Institute for Science and Education (ORISE), Kyle has been at BTS since June 2018, when he completed his master’s degree in geography at the University of Arizona in Tucson. He works on projects that collect, analyze, and visualize big data--namely GPS data from vehicles (e.g., trucks, passenger cars, ships, and aircraft)--to better understand the United States’ national intermodal transportation system and its spatial extent. He is interested in learning more about the role of visualization, especially web cartography and app development, in the effective communication of transportation data and analyses.
Interactive Pedestrian Fatality Web Application
Description
Abstract:
U.S. Pedestrian traffic fatalities have grown both in number and as a percentage of all traffic fatalities in recent years. Addressing this phenomenon is becoming increasingly important. To that end, U.S. Department of Transportation (USDOT) conducted research and developed a predictive pedestrian fatality model using numerous different data sources as part of the USDOT’s Safety Data Initiative . This model quantified risk at the census tract level, and was published in Accident Analysis & Prevention , a copy of which can be found on the Safety Data Initiative Website. While these insights were valuable, we struggled to successfully convey the information in a user-friendly and accessible format to State and local practitioners, policy makers, and the general public. The goal of this data visualization project was to convert the modeling output into a more user friendly format to convey complex results and serve as a network safety diagnostic tool. In pursuit of this objective, we created an interactive web application through the ArcGIS online platform . This Map has an interface to visualize pedestrian fatality risk, and can be filtered by specific parameters relevant to the user such as urban/rural, and risk level. Using intuitive colors and easy to understand controls, a user can zoom into a particular location and view different geographic regions in the U.S. to examine the pedestrian fatality risk results. The information is overlaid on a base map to help users identify specific areas of interest. The map further allows users to explore the data by clicking on individual Census tracks to gather more information about each tract. This approach blends the simplicity of the basic risk ranking information with supporting data elements within the model itself. This project successfully made important transportation safety data more accessible and expanded USDOT’s use of cutting edge visualization software. We believe that this application brings complex modeling into the public realm, and provides an opportunity for State and local practitioners – as well as the public – to use the information as a decision support tool.