Citation

Jiang M, Sato H, Diao X, Mothafer GIMA, Yamamoto T. Transp. Res. Rec. 2023; 2677(5): 1402-1414.

Copyright

(Copyright © 2023, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.1177/03611981221143378

PMID

unavailable

Abstract

This study investigates which factors can influence bicycle accidents for a variety of age groups at the traffic analysis zone level. The cyclists were divided into four age groups: child, young, adult, and elderly. To accommodate for the potential variation in the influence of exogenous factors for these age groups, we utilized the multivariate Poisson gamma mixture model. The formulated model not only incorporates the heterogeneity and over-dispersion of the unobserved factors but also allows for the variance of the covariance between the age groups among the zones. The zone-level variables considered in this study include sociodemographic characteristics, socioeconomic characteristics, the built environment, road network characteristics, and traffic characteristics. The model was estimated using bicycle accident data for the year 2011 in Nagoya, Japan. The estimated results clarify the impacts of the explanatory variables for the respective age groups. Children tend to experience more accidents in their residential zones, while the young and adult age groups have more accidents in zones that have many companies. All age groups are more likely to have accidents in the area of many shops. The elderly group is greatly affected by both the number of bicycle trips and its spatial spillover effect, implying that they are most prone to bicycle accidents among all age groups. The findings of this research can assist decision-makers in making proactive decisions to improve bicycle safety.

Language: en