Risk Factor Analysis of Bicycle Accidents Considering Geometric Features and Bicycle Road at Intersections

Abstract

** 非会員 名古屋大学未来社会創造機構 (Institute of innovation for Future Society, Nagoya University) *** 正会員 名古屋大学未来社会創造機構 (Institute of innovation for Future Society, Nagoya University) **** 非会員 AlpacaJapan 株式会社 (AlpacaJapan Co.,Ltd.) ***** 正会員 名古屋大学未来材料・システム研究所 (Institute of Materials and Systems for Sustainability, Nagoya University) In recent years, the popularity of bicycles as a transport mode has significantly increased due to its numerous advantages. Cycling not only improves the livability of cities, public health and air quality, but also creates hard economic value and jobs (Blondiau et al., 2016). Those characteristics have attracted public authorities’ attention, which led to the increase of bicycle related policies with purpose of increasing bicycle ownership and bicycle trips overall. Japan achieved the third highest bicycle mode share with 12% (2015) after the Netherlands (27%, 2018) and Denmark (19%, 2018) (Takasago et al., 2018), without investing in extensive networks of dedicated cycling infrastructure. However, Japan still has the second highest bicycle fatalities per 1,000,000 persons in 2015 with 6.0 fatalities/1,000,000 person, just after the Netherlands (6.3, 2015) (Takasago et al., 2018). The safety of cyclists is the responsibility of multiple parties, including the user and other travelers, but also transportation planners and engineers through facility design. To attract more bicycle users and to tackle cyclist’s safety issues, Japan has been, in recent years, investing in bicycle infrastructure such as bicycle roads and bicycle lanes. Researchers have focused on revealing bicycle risk factors and analyzing appropriate infrastructure designs and countermeasures that enhance bicycle safety (DiGioia et al., 2017). Diao (2020) investigated zonal bicycle accident risk factors in Nagoya city and found that Person Trip (PT) zones with bicycle facilities show a positive effect on elder group cyclist accidents. However, a majority of road traffic accidents occur at intersections, a complex urban space where different roadway types meet and where conflicts among roadway users are not rare. Because of which, many researchers have focused on the influence of intersection geometric attributes on road traffic accidents to gain a better understanding of the cause of accidents to propose better urban planning and designing of intersections in order to ensure the safety of road users. Numerous studies conducted in Japan have focused on identifying bicycle accident risk factors linked to bicycle users and motorized vehicle behaviors. Yazawa and Kin (2015) investigated the impact of road surface during the day and night time. Nakashima et al. (2014) studied the difference in behaviors between sport bicycle users and city bicycle users. Suzuki and Shimura (2018) and Shimada and Inoue (2018) investigated bicycle users’ behaviors at intersections. Suzuki and Shimura focused on large intersections and found that the crosswalk setback distance and the time elapsed when a cyclist enters the conflict zone are related to the conflict severity between bicycle users and turning vehicles. Shimada and Inoue (2018) investigated bicycle accident risk factors at intersections equipped with bicycle lanes considering the behaviors of bicycle users and motorized vehicles at three intersections in Aichi Prefecture and found that the conflict scenarios between bicycle users and motorized vehicles at intersections are dependent on the road users’ movements, and traffic conditions. A few studies conducted in Japan have focused on intersection geometric attributes impact on bicycle accidents. Shiomi et al. found in 2017 in a study conducted in Aichi, Shiga and Kagawa regions that the intersection size, the crosswalk length and the crosswalk setback distance significantly influence the risk of all types of road traffic accidents, meaning that a compact intersection has a lower risk overall. The difficulty in accessing traffic road accident data and the lack of intersection geometric data represent a major challenge in investigating the impact of intersection geometric features and