MODELLING OF VEHICLE ARRIVALS AT UNCONTROLLED INTERSECTIONS

Abstract

in India, the traffic is mostly mixed in nature, with wide variations in vehicular and traffic characteristics. With the growing traffic on Indian roads, the complexity of flow has increased at uncontrolled intersections, where no priority rules are followed by the users. The vehicle arrival characteristics are one of the basic components of traffic flow modelling. They are the basic requirement in the delay and capacity analysis procedure of two-way stop-controlled or all-way stop-controlled intersections. The models developed to analyze the vehicle arrival characteristics or headway distributions, are mostly for the homogeneous traffic conditions. These models need to be validated for heterogeneous traffic conditions. Thus, an attempt has been made in the present work to model the vehicle arrival patterns observed at selected approaches of uncontrolled intersections in mixed traffic conditions. The data were collected at 18 approaches of uncontrolled intersections in Delhi, Lucknow, Roorkee, Trivendrum, and Warrangal. The data covers a wide range of traffic volume ranging from 120 vehicles/hr/lane to 2484 vehicles/hr/lane, which gives a more generalized view to the problem. The analysis shows that binomial distribution fits the vehicle arrival pattern when coefficient of variation is less than 1, irrespective of the traffic volume. If the value of coefficient of variation is between 1 and 1.3 inclusive of both values then Poisson distribution defines the vehicle arrival pattern and for coefficient of variation exceeding 1.3, negative binomial distribution fits the vehicle arrivals at uncontrolled intersections. When the value of coefficient of variation is 1.3, both the Poisson distribution and negative binomial distribution fit the observed data, indicating a transition state. The present study shows that modelling of vehicle arrivals can be done on the basis of coefficient of variation and the knowledge of complete pattern in terms of mean rate of arrival and variance of arrivals is important to model the vehicle arrivals. It cannot be done on the basis of volume only.