DESIGN OF OVERTAKING SIGHT DISTANCE IN HIGHWAYS USING PROBABILISTIC APPROACH

Abstract

The design of Overtaking Sight Distance (OSD) is an important aspect of highway geometrics. The current procedure of OSD design is based on deterministic methodology. Literatures available report that significant variability is associated with input parameters for OSD design. The present study deals with the characterisation of OSD variability considering different cases of input parameter variability. The Monte Carlo Simulation (MCS) technique has been adopted to derive the distributions of OSD, which are also statistically established by Chi-square Goodness of Fit (GOF) test. In all the cases, OSD is found to follow lognormal distribution. The Coefficient of Variation (COV) of OSD is found to vary from 21.45% to 22.91% and 23.66% to 25.66% for one-way traffic and two-way traffic condition respectively. Thus, the variability associated with OSD appears to be significant and necessitates probabilistic approach in OSD design. A simple design methodology has been presented for reliability based OSD design taking input parameter variability into account.