Truck Accident Severity in New York City: An Investigation of the Spatiotemporal Effects and Vehicle Weight

This paper uses a random parameter spatial ordered probit model to study the truck accident severity in New York City, accounting for spatial dependency, time of day effect, and the heterogeneous effect of truck weight. The sample is drawn from the state maintained incident data, the publicly available Smart Location Data, and the BEST Practices Model (BPM) data from 2010, and includes a variety of accident characteristics, driver and vehicle characteristics, built environment factors, and traffic volume variables. The results show that heterogeneity does exist in the truck weight, but individual crashes are spatially isolated events. The time of day effect, after other environmental factors are controlled, is also insignificant.