Headway Analysis Using Automated Sensor Data under Indian Traffic Conditions

Headway is a microscopic parameter of traffic flow and is defined as the temporal or spatial distance between two consecutive vehicles. On a macroscopic level these headways translate to density and flow, which are two of the fundamental traffic flow parameters. Several studies were reported on headway patterns and the distribution followed by headways, mainly under the homogeneous and lane based traffic. Definition and measurement of these parameters under traffic condition as in India, with lack of lane discipline and heterogeneity in vehicle composition, is a challenging task. Measurement of these microscopic parameters is not easy and hence not many studies reported such analysis under Indian conditions. The present study reports such a statistical analysis of headways on a suburban arterial road in Chennai, using the data collected from a location based automated sensor. Analysis was carried out on the traffic as a whole as well on mode wise characteristics. Data were separated according to the class of leader and follower vehicle and statistical analysis was carried out separately for each class combination. It was found that the average headway of the stream as a whole was in the range of 2.2 to 3 sec. Headway is in the upper range, when heavy vehicles are involved in the leader follower pair. Log-likelihood method was employed to fit statistical distribution to the data. It was found that for all the categories, Weibull distribution is the best fit.