Issue link: http://uwashington.uberflip.com/i/903202
23 2014-2015 Annual Report • Project: Development of Surface-Mounted Smart Piezoelectric Modules for Bridge Damage Identification and Safety Monitoring • PI: Pizhong Qiao (UW), qiao@wsu.edu There is a pressing need to develop an effective safety or health monitoring technique to assess the condition and premature failure of highway bridge structures during their service life and collect useful information and data for bridge repair, maintenance and decision-making so that the economic and human life loss can be avoided. There are many nondestructive methods for inspecting concrete structures, such as radiography, acoustic emission, visual inspection, thermal field, etc.; but the limitations of these techniques, including accuracy, cost, maneuverability, in situ capability, implementation, etc., make them difficult and/or incapable of being applied to in situ and real time structural health monitoring. The objective of the proposed study is to develop an effective non-destructive ultrasonic smart piezoelectric module to be used for identifying the damage and condition (cracks, material degradation, etc.) in highway bridges. Such a smart sensing technology can be used to identify damage in bridge structures, monitor safety conditions, assist bridge maintenance decision- making, help state DOTs perform forensic studies on the bridge premature failure, and meet the PacTrans theme of "developing data driven solutions and decision-making for safe transport". • Project: Safe from Crime at Location-Specific Transit Facilities • PI: Anne Vernez Moudon (UW), moudon@u.washington.edu The proposed research will integrate four data sets in order to provide a state-of-the-art system for monitoring crime in both space and time, and for developing and testing countermeasures for crime prevention. First, we will use location-specific crime data, which several of the region's cities are now making available to the public. These data enable matching crime events with specific transit facilities locations. Second, we will use ORCA card transaction records, which for the first time provide detailed times and locations when transit riders access transit facilities and vehicles, along with the locations and durations of transit transfers. Both crime and ridership data are geolocated and time-stamped, which allow for precise spatial and temporal matching. Third, land use data will bring complementary information on development patterns (residential and employment densities, socioeconomic characteristics of the areas) and activities surrounding transit facilities. Fourth, we will use data on transit stop or station characteristics (shelters, benches, lighting, etc.) to complement land use with micro environment data. The present project will develop and pilot data-driven tools to (1) identify hot spots of criminal activity near and around each transit facility, and (2) assist decision-making regarding the selection of countermeasures and the allocation of future safety investments, using the results of models estimating environmental and socioeconomic predictors of crime near transit facilities. • Project: Developing a Cost-Effective Bus-to-Pedestrian Near-Miss Detection Method using Onboard Video Camera Data • PI: Yinhai Wang (UW), yinhai@uw.edu Public transit and pedestrian safety has gained increasing attention. Bus-to-pedestrian collisions often result in injuries, fatalities and insurance losses. For example, according to Washington State Transit Insurance Pool (WSTIP), a large portion of the collision-related transit losses are with pedestrians in Washington region. However, for the purpose of safety evaluation for a single region, a road segment, or an intersection, the total number of bus-to-pedestrian collisions is commonly not sufficient to support a significant statistical analysis on safety performance evaluation. Near-miss events, on the one hand, are conflicts that need sudden evasive action and has the potential to develop into collisions; on the other hand, they are not commonly recorded due to the lack of detection methods. Thus, developing a cost-effective method to collect near-miss data is necessary in order to identify high-risk locations, scenarios and behaviors