Effective Superelevation on Horizontal Curves of Two lane Rural Road

The aim of this study is to analyze effective superelevation on horizontal curve and number of crashes in two lane rural road. Research carried out in Iran and some of the developing countries indicated that 50% to 60% of crashes in the suburban areas occurred on two lane rural roads. More than half of these crashes happened along the arches and 20% took place along the horizontal curves. The data on these routes were classified into five different dataset categories namely field road inventory, traffic volume, horizontal curve, cross sections and related crash data for Kohgiluyeh Boyerahmad. Then the crash dataset on horizontal curves were abstracted and their effective variables were calculated using Statistical Package for Social Science (SPSS) and Microsoft Excel (MS EXCEL) software. The crash dataset provide comprehensive information about each crash recorded on horizontal curves of the Kohgiluyeh Boyerahmad’s road network. By applying Data Analysis and Statistical Software (STATA) and SPSS to the crash dataset the most significant independent variables were identified namely radius horizontal curve (Rc), deflection angle curve (Δc), superelevation curve (Ec), and average daily traffic (ADT) and these independent variables were used for crash prediction modelling. Nine potential crash prediction models on horizontal curves (CPMHC) on two lane rural roads were developed and the optimum model is CPMHC6 = Exp [(-2.1760 - 0.000279 (Rc) + 0.010474 (Δc) + 0.101199 (Ec) + 0.000164 (ADT)]. The priority and relative importance of these variables were ADT, Ec, Δc, and Rc respectively. However, the design and crash rate reduction for horizontal curves on two lane rural roads were recommended the variable values for maximum Ec, would be equal to 5%.