The analysis of the traffic accident patterns was carried across frequency of accident, speed limit where accident occurred, and the zone that the accident occurred. By using a kernel density smoothing method in combination hypothetically increasing the number of accidents leads to the predictive model of accident likelihood in urban Fairfax county.
Conclusions showed a strong correlation between accidents in residential zones. Speed limit analysis showed that up to 800 accidents sections between 35-45 mph held the most accidents; past 800 accidents the sections there were <=25 mph saw the majority of accidents. From this the County of Fairfax has been able to reduce urban hazards as well as be more prepared for inclement weather.
Medina, R. M., Cervone, G., & Waters, N. M. (2017). Characterizing and Predicting Traffic Accidents in Extreme Weather Environments. Professional Geographer, 69(1), 126-137