Entry Capacity Quantifying Model through Drivers’ Behavior at Roundabouts

Abstract

The prime objective of this study is to develop the entry capacity model for roundaboutsby employing Multiple Linear Regression (MLR) analysis under heterogeneous traffic flow conditions. Required data were collected from 27 selected roundabouts spanning across 8 states of India by using high-definition video (HD) cameras. To reflect the driver behaviour, critical gap and follow up time are estimated by using INAGA method. To develop the robust model, data such as speed of vehicles and lateral clearance to Central Island are collected in microscopically way. To ascertain the lateral clearance of vehicles w.r.t. Central Island of roundabouts, path analysis is carried out by employing Cam Shift algorithm in trajectory analysis. The coefficient of determination (R2) and Nash-Sutcliffe model efficiency coefficient (E) are found to be (0.83, 0.84) and (0.9, 0.91) respectively. The p-value in the test of Analysis of variance (ANOVA) is found to be 0.00 (p-value<0.05) that indicates the proposed model is statistically fit at 95% confidence level. The percentage contribution of each independent variable is observed through sensitivity analysis in this study. It is observed that the variables like lateral distance and speed contributes the most and least variable and sharing about 32% and 11%in the proposed model respectively. These findings will be useful for traffic planners and designers in the capacity estimation of roundabouts under heterogeneous traffic conditions in developing countries with similar traffic characteristics as India.