Simulation Of Highly Heterogeneous Traffic Flow Characteristics

Arasan, V. Thamizh; Dhivya, G. · 2010 · Crossref

DOI: 10.7148/2010-0081-0087

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of simulating heterogeneous traffic flow, a condition prevalent in India where mixed vehicle types operate without strict lane discipline. Existing simulation models (e.g., CORSIM, VISSIM) are designed for homogeneous traffic and fail to capture the complex interactions of heterogeneous streams. To fill this gap, the authors utilize **HETEROSIM**, a microscopic, dynamic, stochastic discrete-event simulation model previously developed by the authors. The study aims to validate this model and apply it to measure traffic concentration, specifically proposing a new metric called **“area-occupancy”** to replace traditional density and occupancy measures, which are ill-suited for lane-less traffic. The methodology involves a two-phase approach: model validation and application. For validation, field data was collected via digital video recording on a four-lane divided highway in Chennai, India. The dataset included traffic composition (51% cars, 16% two-wheelers, etc.), vehicle dimensions, free speeds, and acceleration/deceleration rates for seven vehicle categories. The simulation modeled a 1,400-meter road stretch, incorporating stochastic vehicle generation and movement logic that accounts for lateral clearances and overtaking maneuvers. The model was run with three random number seeds to ensure statistical stability. For the application phase, the validated model simulated heterogeneous traffic on a six-lane road to analyze the relationship between flow, speed, and area-occupancy. The results demonstrate that HETEROSIM accurately replicates field-observed traffic behavior. A comparison of simulated and observed mean speeds for all vehicle categories showed no significant difference, confirmed by a t-test statistic (t=0.89) well below the critical value (2.97) at a 95% confidence level. Furthermore, the study validated the concept of area-occupancy by first applying it to homogeneous (cars-only) traffic, where it exhibited trends identical to traditional density-speed-flow relationships. When applied to heterogeneous traffic, the model generated strong correlations between area-occupancy, stream speed, and flow. Specifically, the relationship between area-occupancy and speed yielded an R² of 0.9948, and the relationship between area-occupancy and flow yielded an R² of 0.9948, confirming the metric's reliability. The significance of this work lies in providing a robust tool for analyzing heterogeneous traffic systems, which constitute a major portion of global traffic in developing nations. By validating HETEROSIM and establishing area-occupancy as a meaningful measure of concentration, the paper offers a framework for evaluating road space usage in mixed-traffic environments. This approach allows researchers and engineers to quantify traffic characteristics that traditional metrics cannot accurately capture, facilitating better infrastructure planning and traffic management in regions lacking lane discipline.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• traffic density