2017 Volume 73 Issue 5 Pages I_1201-I_1209
This study makes a factor analysis of the traffic accident risk in 2 regions, Tokyo and Ehime prefecture. Comparison analyses of the factors are carried out considering the land uses, which affect the rear-end collision accident risk. Region, land use, hourly traffic volume and road structure are considered as the factors, Poisson regression analysis and Negative Binomial regression analysis are carried out using the Road Traffic Census in 2010 and the traffic accident records in 3 years from 2012 to 2014. As a result, it is shown that the impacts of some same factors in Tokyo and Ehime are significantly differential effect of factors. For example, the rear-end collision accident risk can be increased when the level of traffic volume is highest (more than 800 vehicles/h/lane) at DID district in Ehime prefecture, and the traffic accident risk at DID district in Tokyo is decreased at the same level as the hourly traffic volume. On the other hand, the traffic accident risk at DID district in Tokyo is increased as the density of signalized intersection increase and the traffic accident risk at the other district in Ehime prefecture can be increased in accordance with the number of signalized intersection per kilometer increased.