Seasonal crash prediction model for urban signalized intersections: Wisconsin southeast region.

Objectives: Highly aggregated data is conventionally used in transportation safety omitting seasonal variations and leading to important loss of information. States in the northern region of the United States experience significant weather variations with snowfall and ice events. Crash occurrence is the highest during the winter compared to other seasons.