Car Travel Time Estimation near a Bus Stop with Non-motorized Vehicles

Xiaobao, Yang; Huan, Mei; Hongwei, Guo; Liang, Gao Feng · 2011 · OpenAlex-citations

DOI: 10.1080/18756891.2011.9727885

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 estimating car travel time near bus stops in developing countries, where mixed traffic flows involving buses, cars, and non-motorized vehicles (primarily bicycles) create complex interactions and safety risks. While existing traffic models often fail to account for these specific mixed-traffic conditions, real-time travel time estimation is critical for Intelligent Transportation Systems. The study aims to develop a predictive model that quantifies how various traffic factors influence car travel time distributions at curbside bus stops, a common infrastructure type in Chinese cities. To achieve this, the authors employed a hazard-based duration model, specifically the Cox proportional hazard model, which is widely used in survival analysis to determine causality in duration data. Data were collected via video cameras at a curbside stop on Xueyuan Road in Beijing during January 2008, capturing 531 car travel time samples under both peak and off-peak conditions. The study identified four key explanatory variables through data mining techniques: car volume ($X_c$), non-motorized vehicle volume ($X_n$), bus departure volume ($X_b$), and the free ratio of the bus stop ($X_f$). The model estimated the instantaneous probability (hazard rate) of a car passing the bus stop, allowing for the analysis of how these variables modify travel time distributions. The results demonstrated that the proposed model provided a strong fit to the observed data, with a likelihood ratio statistic of 5,471.4. All four variables were statistically significant. The analysis revealed that car volume, non-motorized volume, and bus departure volume negatively affect the hazard rate, meaning higher volumes of these vehicles increase car travel time. For instance, a car volume of 30 vehicles resulted in a travel time 2.25 times longer than at a volume of 10 vehicles. Conversely, the free ratio of the bus stop had a positive effect on the hazard rate; a higher free ratio (indicating fewer stopped buses) reduced travel time. Specifically, the average travel time decreased from 17.35 seconds to 12.57 seconds when the free ratio increased from below 50% to above 50%. The model successfully predicted travel time distributions under varying conditions, such as showing a 17.6% increase in travel time when non-motorized volume rose from 20 to 35 vehicles per minute. The significance of this work lies in its application of survival analysis to transportation engineering, providing a robust method for quantifying the impact of mixed traffic flows on travel time. The findings highlight that bus stop design and traffic management in developing countries must account for the conflicts between motorized and non-motorized vehicles. By offering a tool to estimate travel time under assumed conditions, the model aids in the planning and design of traffic facilities to improve safety and efficiency. The authors suggest future research should incorporate additional factors, such as passenger loads and different bus stop types, to further refine these estimates.

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