Travel Time Estimation Modelling under Heterogeneous Traffic: A Case Study of Urban Traffic Corridor in Surat, India

Saw, Krishna; Das, Aathira K.; Katti, Bhimaji K.; Joshi, Gaurang J. · 2018 · Crossref

DOI: 10.3311/pptr.10847

archive: archived pipeline: cataloged verified

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

Summary

This study addresses the challenge of estimating travel times on urban arterial roads in India, where traffic is characterized by high heterogeneity and recurrent congestion. The research focuses on the Udhana–Sachin Corridor in Surat, India, aiming to identify key attributes affecting travel time—specifically traffic composition, roadside friction, and intersection factors—and to develop a robust estimation model. The motivation stems from the degradation of traffic quality due to vehicle growth, pedestrian encroachments, and haphazard parking, which increase delays and commuter stress. The methodology involved a four-stage approach centered on field surveys conducted during morning and evening peak and off-peak hours. Researchers utilized videography to capture traffic volume and composition, dividing vehicles into four categories: two-wheelers (2Ws), three-wheelers (3Ws), four-wheelers (4Ws), and commercial vehicles (CVs). Speed profiles were recorded using GPS-laden probe vehicles (V-Box survey) to determine spot-to-spot speeds. Roadside friction (RSF), encompassing parked vehicles and pedestrian interruptions, was rated on a scale of 1 to 5 by trained enumerators and normalized to a 0–1 scale. Intersection factors (IF) were calculated as the ratio of average speed to speed drop at uncontrolled intersections. These data were analyzed segment-wise across three 2 km sections of the 5.8 km corridor. The findings reveal that Segment I, characterized by high commercial and residential density, experienced the highest traffic volume (approx. 5900 vehicles/hour during peak) and the longest travel times. Traffic composition was dominated by 2Ws (53–60%) and 3Ws (28%). A Multi-Linear Regression (MLR) model, termed TRATIM, was developed with an R² of 0.82. The model equation indicates that 3Ws, CVs, RSF, and IF positively correlate with travel time, while 4Ws exhibit a negative correlation, likely because probe vehicles follow faster four-wheeler platoons. Model validation showed a Root Mean Square Error of 0.18 minutes and a Mean Absolute Percentage Error of 7.02%. Sensitivity analysis confirmed that travel time increases with higher volumes of 2Ws, 3Ws, and CVs, but decreases with more 4Ws. Additionally, increases in RSF and IF significantly impede speed, with IF having a slightly higher impact than RSF. The study concludes that heterogeneous traffic composition, alongside roadside friction and intersection characteristics, are critical determinants of travel time in Indian urban corridors. The developed TRATIM model provides a reliable tool for transportation planning and traffic management. The authors recommend mitigation strategies such as restricting heavy vehicles, implementing access control at high-impact intersections, and regulating parking and pedestrian movements to improve corridor performance.

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.

• exposure measurement