Safety Modelling of Urban Road a Case Study of Tiruchirappalli (Presentation 889)

Road safety is the main concern in many developing countries including India. According to Ministry of Road Transport and Highways, Tamilnadu reported the maximum fatalities in the past year, accounting for about 10.7% of total fatalities in India. Rapid motorization accompanied by quick expansion in road network and urbanization seriously impacts the road safety levels of our country. In India, road accident is the major disaster that has taken 17 lives every hour over the past year. Lack of scientific analysis of accidents and non-adherence of general safety measures lead to undesirable results. Traffic safety on urban arterials is influenced by several key variables including geometric design features, land use, traffic volume, and travel speeds. This study aims to analyze the relationships among these variables. The classified accident data for six years (2010 to 2015) were collected from Tiruchirappalli Commissioner’s Office and Police Records for urban arterials of Tiruchirappalli city. Geometric data was obtained from field investigations. Volume studies and spot speed studies were conducted on the study segments. Poisson-lognormal and Negative Binomial model was used to estimate the accident frequencies. The results show that Negative Binomial models had the best goodness- of-fit and efficiency with the lowest standard errors. The models explain that traffic volume, speed and geometric design features significantly affect the accident occurrence.