University of Washington

21 2017 – 2018 Annual Report Understanding Freight Behavior in the Pacifi c Northwest: An Evaluation and Application of EROAD Data to Freight Demand and Forecast Modeling • PI: Salvador Hernandez (OSU) In summary, this research's objective will be achieved through the application of a private data source acquired as part of an ongoing study with EROAD. EROAD is a company that develops and implements technology to modernize traditional paper-based systems within the trucking industry. As part of this modernization, EROAD collects the data used for modeling and forecasting freight movements. However, EROAD data has yet to be used for such an application. This will be accomplished through (1) a comprehensive existing state-of-the-art and state-of-the-practice as it relates to freight data, both in terms of analysis and collection, (2) a massage of EROAD data and conduct statistical analyses for freight movement analysis, and (3) identify freight movements and detailed supply-chain analyses of key commodities. It's envisaged that two journal manuscripts will result from this work. Bridge Structural Inspections using Bridge Information Models (BrIM) and Unmanned Aerial Vehicles (UAVs) • PI: Yelda Turkan (OSU) This research project will develop a novel framework that implements Bridge Information Modeling (BrIM) and Unmanned Aerial Vehicle (UAV) technologies to improve current manual bridge inspection technique. UAVs enable safe and rapid collection of visual bridge inspection data in the form of digital images. BrIM, on the other hand, enables storing all bridge data, including its drawings and 3D models, material specifi cations, inspection notes, images and others, in a central object-oriented database that can be accessed from both the offi ce and the fi eld. Computer vision algorithms will be used to detect cracks from UAV images automatically based on statistical learning theory. This information will be used to automate the creation or update of bridge database, i.e. 3D BrIM. Using the proposed framework would enable automated and remote assessment of bridge conditions. The framework will implement the state-of-the-art condition assessment codes and standards, and it should contribute to a) providing an improved means of accurately documenting the structural condition assessment data, b) eliminating errors resulting from data transcription, and c) providing more cost effective and safer bridge inspections. All of these factors should help assist in maintaining U.S. bridges in a state of good repair. Commuter Mobility Versus Work-Zone Safety: An Empirical Study using Data Mining of Anonymous User Generated Trip Information • PI: Joseph Louis (OSU) The deterioration of the nation's highway system through wear and tear necessitates periodic maintenance activities, which inevitably puts construction workers on work zones in close proximity with the traveling public. While the safety of workers and travelers is paramount in the context of the work zone, it must necessarily be balanced with the mobility of the traveler. Thus the goals of safety and mobility, which are often spoken of in the same breath when it comes to highway work zones, are often in confl ict with each other. This research proposes to study the relationship and the tradeoffs between safety measures on highway work zones and the effect that they have on the mobility of the traveling public. Specifi cally, the travel times between two sections of the highway will be studied through the analysis of crowd sourced data collected from mapping services providers like Google Maps and Waze both before and after the introduction of a work zone. This will enable the quantifi cation of delay that occurs as a result of the work zone, which can then be correlated with the different safety measures implemented on site. It is expected that this research will improve our understanding of the safety-mobility balance on highway work zones and enable state agencies and contractors to optimize their safety plans without unduly burdening the traveling public.

• Cover