Development and Application of Accident Prediction Model for Intersections at Dhaka City

Abstract

Pedestrians are some of the most vulnerable road users, especially in large congested cities in developing countries. Commencing with the data that every year more than 1.3 million people die in road accidents around the world (Building Leaders in Urban Transport Planning, LUTP, 2012). So, this indicates that the major mortality rate is procured by the road accidents no question asked. Not only in USA or any other developed countries of the world, but also in developing country like Bangladesh, with regard to safety, the pedestrians are of major concern as they represent up to 72 percent of road traffic fatalities in Dhaka Metropolitan City (Rahman et al., 2006). So, it can be stated that pedestrian safety is not only a problem in global range, but also acute in developing country like Bangladesh. In order to develop appropriate countermeasures to improve safety, research has to be conducted to understand the factors contributing to vehicle-pedestrian collisions. This study is especially designed to predict accident probability only for pedestrian-vehicle crashes at intersections as accurate as possible. The statistical model developed in this study are focused in predicting traffic accidents at four-legged signalized and tee or three legged intersections. For data analyzing and modeling of the pedestrian-vehicle crashes at intersection, at first annual crash data were collected from 1998-2009 from ARI (Accident Research Institute), BUET. There is a total of 54 intersection’s data were available and we worked on 45 selected intersections. After getting all the data analyzed by the statistical software, a projected equation is possible to make with the help of the critical and significant factors to get accident prediction for accident forecasting at any particular intersection. Using data from the capital of Bangladesh, Dhaka, this study finds that waste deposit facilities, overhead bridges and underpasses, solar panel system, law enforcement authority and vehicle volume are associated with an increase in the number of vehicle-pedestrian crashes, whereas roads with the rail crossings, similar approach road, commercial land uses, maximum number of inbound and outbound lanes, roads with greater number of lanes, absence of bus stop, speed breakers and pedestrian volume are associated with a reduction in the number of vehicle-pedestrian crashes.