Entry capacity modelling of signalized roundabouts under heterogeneous traffic conditions

The primary objectives of this study are to develop two signalized-based roundabouts entry capacity model by employing regression-based multiple non-linear regression model (MNLR) and artificial intelligence-based age-layered population structure genetic programming (ALPS GP) model under heterogeneous traffic conditions. Based on the modified rank index (MRI) values, the ALPS GP model is found to be most suitable model under heterogeneous traffic conditions. However, in a practical point of view, MNLR-based signalized model is recommended for determining roundabout entry capacity under heterogeneous traffic conditions. Sensitivity analysis reports that weaving length is the prime variable and sharing about 27.72 % in the MNLR-based signalized roundabout entry capacity model. These findings will be useful for traffic planners and designers in the capacity estimation of signalized roundabouts under heterogeneous traffic conditions in developing countries with similar traffic characteristics as India.