1 ESTIMATING TRUCK FLOWS ON A FREEWAY NETWORK TO IMPROVE FREIGHT OPERATIONS AND PLANNING Robert L. Bertini 1* , Kristin Tufte 1 , Zachary Horowitz 1 and Spicer Matthews 1 1. Department of Civil and Environmental Engineering Portland State University P.O. Box 751, Portland, Oregon 97207-0751 USA Phone: 503-725-4285 Fax: 503-725-5950 Email: bertini@pdx.edu ABSTRACT Increased knowledge of truck travel patterns has the potential to increase overall highway safety, lead to better-managed maintenance operations, provide cost savings to public agencies, validate investments in intelligent transportation systems, and improve long-range planning and forecasting. This paper explores techniques that use current ITS technologies such as video image processing and loop detector data algorithms to collect and verify short and long vehicle count and length data. Three sets of traffic data for each time interval were created, and then compared using statistical analyses to produce results that that reveal some new information about the freight transportation system in the Portland metropolitan region. KEYWORDS freight, archived data, traffic management INTRODUCTION The growing volume of freight that is increasingly moved by trucks on the nation’s transportation network makes it more important than ever to develop a better understanding of the freight system’s overall operational characteristics. This project attempts to establish a base level of knowledge regarding the characteristics of the movement of freight in the Portland region through the analysis of long vehicle (LV) volumes and speeds. A body of knowledge about the technical aspects of freight data collection has been established through experimentation with ITS subsystems such as the CCTV network, video image processing technologies, inductive loop detectors, the traffic data storage networks, and data processing algorithms. This project had several objectives. First, the ability of researchers and public agencies to collect freight data using the existing Oregon Department of Transportation (ODOT) CCTV camera network was tested. This was accomplished by coordinating camera positioning with the ODOT traffic management operation center and the successful connection of the fiber optic data feed to the Intelligent Transportation Systems (ITS) Laboratory at Portland State University. Second, it was demonstrated that traffic recorded on DVDs could be processed with a video image processing system (Autoscope RackVision), and done such that a methodology was established that can be used for future data collection experiments. Third, the robustness of the Nihan-Wang algorithm [1] was established for data collected in the Portland metropolitan region. Other literature was also reviewed [2-6], and discussed more fully in a related research report. [7]