A Spatiotemporal Analysis of Traffic Crashes in Minnesota

Keywords: Traffic crash, spatiotemporal analysis, logistic regression, traffic safety, Minnesota
Session Type: Virtual Paper Abstract
Day: Sunday
Session Start / End Time: 2/27/2022 11:20 AM (Eastern Time (US & Canada)) - 2/27/2022 12:40 PM (Eastern Time (US & Canada))
Room: Virtual 11

Abstract

Using traffic crash data compiled by Minnesota Department of Transportation (MnDOT), this study examined the spatial and temporal patterns of motor vehicle accidents from 2015 to 2019. The results of spatial autocorrelation analysis and hotspot analysis show spatial distribution patterns similar to those in other states with most crashes occurring within or in the close vicinity of urban areas or along highways. Nonetheless, there are strong patterns of spatial autocorrelation even with the exclusion of cities. The results also reveal significant temporal (e.g., seasonal, monthly, weekly, hourly) variations in crash occurrence. Ordinal logistic regression was used to explore the association between crash severity level and a variety of factors that reflect driver personal characteristics, road conditions and environmental characteristics. The results suggest that significant differences exist between factors contributing to injury severity. Young drivers and male drivers tend to be involved in more severe accidents. Poor road conditions (measured by surface type, surface geometry, and surface condition), light conditions and alcohol/drug use contribute to severe crashes. Based on the findings, the study discussed the measures that MnDOT and other agencies may develop to reduce crash frequency and severity and to improve traffic safety in the state of Minnesota.