Development of Optimized Traffic Stream Models Under Heterogeneous Traffic Conditions

Macroscopic traffic stream models provide relationships among the three basic traffic variables namely speed, flow and density for steady-state conditions. A number of research papers are available proposing different traffic stream models under homogeneous traffic conditions. These models may not be directly suitable for the Indian traffic condition which is heterogeneous and less lane disciplined in nature. Not much research has been reported on this important aspect so far from India. The present study is an attempt to develop an optimal traffic stream model suitable for the heterogeneous traffic flow condition existing in India, taking Chennai as a case study. Several existing functional forms were attempted, both single regime and multi regime, before reaching the best fitting two-regime speed-density relationship using a conventional optimization algorithm. Using this, the relationships between speed-flow and flow-density were theoretically derived. The results indicate that the proposed model is able to represent the steady-state macroscopic behavior of the traffic stream with reasonable accuracy.