查看文章

Abstract of the final report is stated below for reference: This project’s ultimate deliverable is a functional Vulnerable User Density Dashboard (https://mti. umd. edu/sdi) for the state of Maryland. The dashboard uses mobile device location data and electric scooter volume data to reflect pedestrian, bicycle, and electric scooter travel volumes and their exposure to roadway safety risk across all roadways in the State. The team develops advanced statistical models to study and predict pedestrian and bicycle involved crashes in a first of its kind effort to generate vulnerable roadway user risk at the link level. The results indicate high correlation between estimated volumes and observed frequency of crashes. This estimated volume and exposure data fills an important gap in understanding the spatial and temporal distributions of pedestrian and bicycle activities. Through this dashboard engineers, planners, and stakeholders can quickly identify safety risk hotspots for vulnerable road users across Maryland and start parsing out why certain locations with high volumes have less crashes. This data-driven dashboard uses emerging data sources and cutting-edge big-data analytics to derive volume estimates and crash risk predictions. It can be used by different stakeholders for situational awareness and traffic analyses of vulnerable users, ie, pedestrians and bicycles. The predicted risk exposures and hotspots will support relevant decisions such as identifying locations for improvements and implementing pedestrian and bicycle safety countermeasures.