نبذة مختصرة : In developing nations like India, motorization is increasing along with economic growth. Road traffic deaths in urban India have consistently been a serious issue of concern. National Crime Records Bureau (NCRB) 2014 reports show that urban road traffic crashes within the state of Kerala, India, increased by 37% from 2009 to 2012. Nearly 20% of those crashes occurred at intersections. 40% of significant traffic-related injuries and fatalities involved incidents at signalized intersections, which made up 24% of all recorded crashes at intersections. An urban road network's signalized crossings are one of its biggest weak points. By collecting six years crash data of signalized intersections of Kollam corporation it is found that signalized intersection crashes are increasing each year crashes increased by 72% within six years. It implies the need to control the crashes occurring at signalized intersections of Kollam corporation. The current study investigates the formation of crash frequency prediction model and crash severity prediction model for signalized intersections of Kollam corporation by doing statistical analysis of the crash data. There are totally ten signalized intersections within the Kollam corporation. Six types of regression models are used to analyze crash frequency and the model which best fit the data is chosen as final prediction model for total crashes and grievous crashes. Ordered probit model is employed to form crash severity prediction model and marginal effects are determined which helps to understand effects of each factors on severity levels. A better understanding of the crash causative factors aids to develop more targeted countermeasures for improving the safety and performance of signalized intersections. Assessment of safety at signalized intersection aid traffic and road safety engineers to adopt better solutions for reducing crashes. Modeling relationship between crash frequency, severity and it’s determining factors help to achieve knowledge about crash occurrence mechanism ...
No Comments.