Identifying crash black spots in Melbourne road network using Kernel Density Estimation in GIS

In safety studies, accident black spot refers to a place with a record of large number of crashes or crashes with high severity. Identifying road accident black spots assist traffic engineers in working on these areas to decrease number of traffic accidents and reduce crash severities. Different approaches exist in the literature to identify the accident black spots. However, road accident black spots must be recognized according to crash statistics and the factors influencing road accidents. From 2008 to 2012, about 9,800 accidents occurred each year in Melbourne metropolitan area. In these accidents, cars (Passenger cars, utilities and vans) were involved in 81% of accidents, and motorcycles and bicycles involved in 7% of accidents. In addition, trucks, buses and trams involved in 3.5%, 1% and 0.4% of accidents, respectively. The aim of this research is to use Geographical Information Systems (GIS) and Kernel Density Estimation (KDE) to find the spatial patterns of accidents in Melbourne metropolitan area for different type of vehicles. This paper also identifies important factors influencing these accidents. Using KDE and other spatial statistic tools enable us to find the crash risk distribution in road networks and allocate our resources to improve safety in these crash hot spots.