Red-light running behavior of delivery-service E-cyclists based on survival analysis

Abstract

Objective

The primary objective of this study is to explore the red-light running behavior of delivery-service E-cyclists, including differences with regular E-cyclists and influencing factors.

Methods

A total of 2173 E-cyclists in Shanghai were observed, with a mix of 51.8% regular E-cyclists and 48.2% delivery-service E-cyclists. Survival analysis was used to establish the model to resolve the issue of censored data of the waiting time of E-cyclists at an intersection. The Kaplan–Meier estimator was adopted to examine the significance of the difference between regular E-cyclists and delivery-service E-cyclists on red-light running behavior. A Cox proportional hazards model with six potential influencing factors was developed to estimate the red-light running probability of delivery-service E-cyclists.

Results

The violation rate of the red-light running behavior is almost 40% higher for delivery-service E-cyclists when compared to that for regular E-cyclists. The results show four factors that increase the hazard rate of red-light violation for delivery-service E-cyclists: being male, visual search (i.e., head movement), waiting beyond the stop line, and existence of red-light running of other (E-)cyclists. Additionally, they show one factor decreases the hazard rate of red-light violation: group size.

Conclusions

Waiting position, violation of the law by other cyclists, and group size play an important role in red-right running behavior. The hazard rates of running red-light by delivery-service E-cyclists increased by 62% and 33% when they wait near motorized lanes and when other individuals violate traffic rules, respectively. The hazard rates reduced by 50% when there are more than five waiting cyclists.

Keywords:

• Red-light running behavior
• E-cyclists
• survival analysis
• intersection

Acknowledgments

The authors appreciate the editor and four anonymous reviewers for the insightful review. Their extremely detailed comments and suggestions help a lot in improving the quality of the paper.

Data availability statement

The datasets generated during the current study are available in the Mendeley Data repository, [http://dx.doi.org/10.17632/s2jgpv8m9f.2].

Additional information

Funding

The research is supported by the National Natural Science Foundation of China under grant No. 71971140.