| Priority setting for road safety improvements is a complex issue. Developing countries like Bangladesh suffer from the lack of road safety resources and there is an obvious need for the effective utilization of these scarce resources. The first part of this thesis describes a simple framework for road safety priority setting for different geographical areas in Bangladesh. Data Envelopment Analysis (DEA) is used to examine the relative risks of different regions and identify high risk regions. We found that the relative collision risks of various regions are not directly proportional to the number of casualties and transportation accessibilities of those corresponding regions as commonly believed. We therefore recommend that systematic decision making methods like DEA be used to identify high risk regions to maximize return of scarce road safety resource. Apart from identification of spatial distribution and concentration of pedestrian accidents, identification of the prominent (statistically significant) risk factors is another way to provide cost effective solutions of road safety problem. In the second part of the thesis, an attempt is made to explore the factors that contribute to mortality and injury, using historical traffic collision records. It is found that factors like pedestrian above the age of 55, young pedestrians age under 15 years of age, pedestrian crossing action, involvement of trucks or buses, pedestrian crosswalk locations, locations without traffic control devices, rainy season are some important pedestrian collision risk determinants. |
