This study analyses the built environment factors that most impact the severity of injuries in cyclist accidents in Lisbon, Portugal. The most common type of bike lane in Lisbon is the protected bike lane, making up 49% of all bike lanes. This is positive from a safety point of view since the physically separated bike lanes offer the best protection for cyclists. On the other hand, of the 24 districts that compose the city of Lisbon, there are 7 with less than 1 km of protected bike lanes, showing that the distribution of the bicycle infrastructures can be improved upon. The data on which this study is based are for the years 2015 to 2019 (inclusive). To determine which built environment is most relevant in injury severity, 2 modeling runs were performed: multinomial logistic regression, and ordinal logistic regression. The dependent variable chosen was “Injury severity”, with 3 distinct categories: unharmed (40 cases), slight (387), and severe (12). It was seen that accidents that occurred at intersections, at night, or in the central zone of Lisbon had a higher probability of resulting in severe injuries than accidents that occurred outside of these conditions.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Characterization of Accidents with Cyclists in Lisbon: Study on the Built Environment Factors

  • Sebastião de Sousa Barreto,
  • Luís Picado-Santos

摘要

This study analyses the built environment factors that most impact the severity of injuries in cyclist accidents in Lisbon, Portugal. The most common type of bike lane in Lisbon is the protected bike lane, making up 49% of all bike lanes. This is positive from a safety point of view since the physically separated bike lanes offer the best protection for cyclists. On the other hand, of the 24 districts that compose the city of Lisbon, there are 7 with less than 1 km of protected bike lanes, showing that the distribution of the bicycle infrastructures can be improved upon. The data on which this study is based are for the years 2015 to 2019 (inclusive). To determine which built environment is most relevant in injury severity, 2 modeling runs were performed: multinomial logistic regression, and ordinal logistic regression. The dependent variable chosen was “Injury severity”, with 3 distinct categories: unharmed (40 cases), slight (387), and severe (12). It was seen that accidents that occurred at intersections, at night, or in the central zone of Lisbon had a higher probability of resulting in severe injuries than accidents that occurred outside of these conditions.