VEHICLE CLASS WISE SPEED-VOLUME MODELS FOR HETEROGENEOUS TRAFFIC

Thomas, Jomy; Srinivasan, Karthik K.; Arasan, Venkatachalam Thamizh · 2012 · Crossref

DOI: 10.3846/16484142.2012.697442

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 inadequacy of standard Volume Delay Functions (VDFs) for modeling traffic in developing countries, where heterogeneous traffic mixes and imperfect lane discipline render homogeneous, lane-based models ineffective. Traditional VDFs aggregate all vehicles into Passenger Car Units (PCUs), which fails to capture asymmetric interactions between vehicle classes and prevents the determination of class-specific travel times. The authors aim to develop vehicle class-specific speed-flow models that account for varying volumes and compositions, thereby providing a more realistic representation of mixed traffic dynamics for applications such as traffic assignment and Level of Service (LOS) analysis. To overcome the limitations of field data collection, the study utilizes HETEROSIM, a calibrated micro-simulation model designed for non-lane-based heterogeneous traffic. The model was validated against field data from six-lane divided urban arterials in Chennai, India, showing acceptable error margins for average speeds. The researchers simulated traffic across a wide range of volumes (from 500 vehicles per hour to capacity) and 25 distinct composition sets representing heavy vehicles, cars, auto-rickshaws, and motorized two-wheelers. This experimental design allowed for the systematic observation of how speed for each vehicle class responds to changes in total volume and the specific mix of vehicle types. The results demonstrate that class-wise speeds vary significantly depending on volume levels and traffic composition, invalidating the use of a single VDF for all vehicle types. At low to moderate volumes, cars exhibit the highest speeds, while two-wheelers and autos are slower; however, this order reverses at high congestion levels, where two-wheelers maintain higher speeds than cars due to their ability to filter through traffic. The study identifies distinct regimes in speed-flow relationships, noting that the sensitivity of speed to volume increases as congestion rises. Furthermore, the models reveal asymmetric interactions, where the presence of certain vehicle classes disproportionately affects the speed of others. The proposed multi-class, multi-regime models significantly outperformed traditional single-class VDFs in both calibration and validation datasets. The significance of this work lies in its provision of a robust framework for analyzing heterogeneous traffic without relying on homogenizing PCU conversions. The models enable the calculation of class-specific LOS, revealing that larger vehicles suffer more adversely from congestion than two-wheelers. Additionally, the study demonstrates the utility of these models for policy analysis, such as evaluating the impact of excluding specific vehicle classes during peak hours. By capturing the differential effects of composition and volume on each vehicle type, the proposed models offer improved accuracy for determining road user costs, emissions, and performance measures in mixed traffic environments.

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