Friday, May 26, 2017

Characterizing and Predicting Traffic Accidents in Extreme Weather Environments.

This article concerns itself with the making of a traffic accident predictive model based on the Doppler data of Fairfax County, VA after an extreme weather event, in this case snowstorms. This will effectively estimate where hot-spots for accidents occur while increasing predictive capabilities with every input of data. Before analyzing the Doppler data, the systematic processes of climate change, urban migration, and aging infrastructure must be accounted for.


Figure 2 Ground Doppler weather radar visualizations of the 2011 Washington, DC, metropolitan area storm. (Color figure available online.)  


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 Geographer69(1), 126-137

1 comment:

  1. I cannot thank lemeridian funding service enough and letting people know how grateful I am for all the assistance that you and your team staff have provided and I look forward to recommending friends and family should they need financial advice or assistance @ 1,9% Rate for Business Loan .Via Contact : . lfdsloans@lemeridianfds.com / lfdsloans@outlook.com. WhatsApp...+ 19893943740. Keep up the great work.
    Thanks, Busarakham.

    ReplyDelete