Development of Roundabout Entry Capacity Model Using INAGA Method for Heterogeneous Traffic Flow Conditions

The main objective of the study is to develop a roundabout entry capacity model by employing Influence Area for Gap Acceptance (INAGA) method under heterogeneous traffic flow conditions. In this regard, data collection has been carried out to extract the traffic flow and gap acceptance parameters which were significantly contributing to modelling the roundabout entry capacity. The traffic flow data were found to be heterogeneous in nature because of the percentage of traffic share in entry leg varying (0–55.8)%, (10–79.5)%, (18–76.5)%, (0–25.5)% and (0–7)% of heavy vehicles, light motor vehicles, motorcycles/scooters, bicycles and human-drawn vehicles, respectively. The critical gap (tc\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t_\mathrm{c}$$\end{document}) and follow-up time (tf\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t_\mathrm{f}$$\end{document}) were found by (INAGA) method ranging between (0.88–2.68) s and (0.86–2.65) s, respectively, via comprised all sites together. Then, multivariate regression analysis was carried out by utilizing M1 distribution function. The analysis of variance demonstrates that the model was statistically significant at 95% confidence level with measured R square is 0.78. To assess the prediction performance of the model, data splitting method was employed in model validation process and revealed that field capacity varies 2.5% w.r.t model capacity in which the measured R square is 0.9758. The newly developed model was compared with different existing models to judge the appropriateness of the model. HCM 2000 and HCM 2010 models showed nearly same results as the developed model. While the Brilon–Wu formula (Germany) and GIRABASE formula (France) overestimated and underestimated the capacity for all flow ranges, respectively.