An improved method for accident estimation and comparison Types of urban level intersections

Growth of population and urbanization, transport expansion as well as demand of road trips in developing countries such as Iran, cause to the increase of the number of vehicles day by day, which this matter leads to the enhancement of the unreasonable of travel time and high rate of accidents. The data of accidents show that in whole of the world around 13 million people die every year from accidents and more than 50 million people are injured or physically handicapped due to this matter. Therefore, in order to achieving the proper traffic planning to reduce of accidents number and casualties, assessing the road safety with considering effective factors is very important. In general, different methods are used to evaluation of the safety of transportation facilities, which most commonly methods of them are statistical models. The main purpose of this paper is to present an improved method, to evaluate the safety of different urban intersections by using a statistical model of urban accidents. This study was conducted on 140 three-way and four-way intersections which included lights and no lights cross. In these cases data of accidents on three years period, from 1387 to 1390, have been collected from the 113 police reports and other required data have been gathered by field investigation (research). A multiplicative form based on Haier equation has been used to build the considered model, where descriptive variables choose in continuous and binary forms and with a predetermined function are introduced to the equation. The results showed that in the urban intersections, incoming traffic has the greatest impact on accident rates, and for a certain traffic volume, effects of intersection geometry (3-way and 4-way) and also type of traffic control (signalized and no lights) on the rate of accident are negligible.