This is an interesting and useful result of an unofficial public-private partnership where the City and County of Honolulu provided a database of redacted accident records with geographic identification data and a private firm used geographic information system (GIS) expertise to provide a depiction and summary of these data by location. The 2015 Traffic Accident Map of Honolulu by the Law firm of Davis Levin Livingston lets one quickly identify traffic black spots.
For example, the portion of their map I captured above shows that the University of Hawaii area is generally light in crashes. Punahou Street near the freeway has a moderate amount of crashes. The set of blocks surrounding and including Ala Moana Cednter, one of the nation's largest shopping centers, is by far Hawaii's largest black spot, although, I guestimate that most of the reported crashes there are of low severity and the area depicted is of relatively low risk.
One must keep in mind , that high accident spots are not necessarily high risk or high danger spots. As you'd expect, locations with high traffic are also high crash and accident spots. Only if we divide the number of crashes by the amount of traffic occurring in a typical day we can get a better representation of risk.
For example, Location A has recorded 1 crash and gets an average daily traffic (ADT) of 10,000 whereas location B has recorded 8 crashes and gets an average daily traffic (ADT) of 100,000. In this case, B has a higher number of crashes but A has a relatively higher risk for crashes.
One must keep in mind , that high accident spots are not necessarily high risk or high danger spots. As you'd expect, locations with high traffic are also high crash and accident spots. Only if we divide the number of crashes by the amount of traffic occurring in a typical day we can get a better representation of risk.
For example, Location A has recorded 1 crash and gets an average daily traffic (ADT) of 10,000 whereas location B has recorded 8 crashes and gets an average daily traffic (ADT) of 100,000. In this case, B has a higher number of crashes but A has a relatively higher risk for crashes.