University of Washington

Research Success Stories DATA COLLECTION AND SPATIAL INTERPOLATION OF BICYCLE AND PEDESTRIAN DATA Public transit and non-motorized travel modes represent convenient, affordable, and sustainable alternatives to personal motor vehicles for many travelers. However, understanding travel patterns and mode choice behavior is crucial to making well- informed planning and operations decisions that will optimize the allocation of limited resources and increase participation in these modes. Toward this end, this project is focused on leveraging current data collection efforts to obtain more complete travel demand information, and introducing new methods that have the potential to reduce cost and improve the quality and coverage of transportation data. First, this study first conducted a poll via online questionnaire to gain a better understanding of how communities are conducting manual bicycle and pedestrian counts, and how the data is used in decision making. A field study was also conducted to evaluate the manual counting errors that arise under different scenarios. Next, a novel methodology for spatial extrapolation of manual and automated mechanical bicycle count data was developed and applied to real-world data from the city of Bellingham, WA. With this approach, it is possible to estimate network-wide bicycle volumes from a relatively small number of discrete count locations. Using this method of network-wide demand estimation, a GIS software tool was developed to assess cyclist accident risk exposure and evaluate various improvement scenarios. Next, this study introduced a set of data acquisition and analysis techniques that apply Bluetooth and Wi-Fi MAC address sensing technologies to estimate origin/ destination (OD) and demand information for non-motorized travelers and transit riders. While stationary wireless sensing has been employed by a number of DOTs, this study takes a novel approach in which mobile sensors are distributed throughout the transportation network. Specifically, a smartphone application was developed to collect Bluetooth and Wi-Fi MAC addresses and GPS location data, and deployed using volunteers in Moscow, Idaho over a 10 day period. Sensors were also installed on four University of Washington (UW) Transportation buses in Seattle, WA over a four week period. Methodologies for data processing, OD estimation, and mode inference are introduced to exclude invalid detections and produce useable information for planning and decision making. The result of this work demonstrates the feasibility of a low-cost and effective data acquisition tool, and received excellent feedback from partners at UW Transportation. Contact: Michael Lowry, mlowry@uidaho.edu On February 10, officials from Washington State Department of Transportation (WSDOT) met with PacTrans to discuss DRIVE Net, the Digital Roadway Interactive Visualization and Evaluation Network. WSDOT expressed interest in establishing DRIVE Net as the organization's data management and analysis tool, and discussed system maintenance options with the STAR Lab research team. WSDOT's Statewide Travel and Collision Data Office (STCDO) learned about data stored in DRIVE Net, and is willing to contribute more data to the system. DRIVE Net is an online platform for transportation data sharing, modeling, visualization, and decision support. Both WSDOT and PacTrans funding were received in developing the system. Contact: Yinhai Wang, yinhai@uw.edu DRIVE NET OFFERS DATA OPPORTUNITIES FOR ORGANIZATIONS 9 2014-2015 Annual Report

• Cover