Fitting Statistical Distributions for Speeds and Headways for Peak and Non-Peak Mixed Traffic Flows in Mangalore City, Karnataka, India

Statistical methods can be employed effectively in analyzing and interpreting observed phenomenon such as traffic arrivals, headways, and vehicular speeds. This paper describes the application of statistical distributions in fitting observed data on speed and headways for specific midblock sections in Mangalore city. As part of this exercise, the relevant data were collected, and attempts were made to fit the normal, log-normal, Erlang, negative exponential and the shifted exponential distributions to the observed traffic data. The best fitting statistical distributions were identified using the Chi-square tests and the Kolmonogrov-Smirnov (K-S) tests for goodness of fit. The findings of this study will further assist researchers in adopting the best approach in modeling urban road traffic especially in mixed traffic conditions comprising various combinations of fast and slow moving vehicles.