Clustering of Vehicles by Performance Characteristics for Indian multi-lane Highways

Vehicle type plays an important role in the accident occurrence, lane changing process and capacity analysis. In heterogeneous traffic, like Indian roads, vehicle interaction is high due to different types of vehicles with difference in speed. Sizes of vehicle-type are same in existing vehicle classification. But, the performance characteristics of vehicle types are different. Grouping the vehicles based on performance characteristics and vehicle configuration is needed for multi-lane highways. The present study is an attempt to identify and group the same performance characteristics vehicles using dynamic cluster based on free-flow speed and Vehicle configuration. Classified traffic volume and spot-speed study have been carried out for a four lane divided highway (NH7 – Virudhunagar to Madurai stretch, Tamilnadu) in India. The road stretch has been chosen as far as away from the city influence to experience the free flow conditions. Considering only free-flow speed, Euclidean lengths for each existing vehicles types have been calculated. Based on Euclidean lengths, vehicles are classified into five groups. This five groups are further modified as eight groups considering the vehicle configuration also. Traffic composition and standard deviation of speeds for new vehicle types have been obtained as the performance characteristics.