.png){kind=icon}
返回
Mapping micro-mobility risk: AI-powered geospatial analysis and predictive modelling☆
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ssci.2025.107107.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Micro-mobility vehicles have rapidly become widespread as a sustainable and practical alternative for urban transportation in recent years. In this study, micro-mobility vehicles refer to traditional bicycles, electric bicycles, and electric scooters, which represent the main categories of such modes involved in traffic crashes in T & uuml;rkiye. Despite their growing popularity, the safety implications of these vehicles have not yet been fully understood, and comprehensive research addressing crash patterns and associated risk factors is required. To this end, this study employs an artificial intelligence-driven geospatial and statistical methodology. Crash reports involving micro-mobility vehicles in T & uuml;rkiye between 2015 and 2023 were analysed. Seventeen independent variables and 102 sub-variables were identified and integrated into a GIS environment for spatial analysis. The impact levels of risk factors were assessed using six different Large Language Models (DeepSeek, GEMINI, Perplexity, ChatGPT, Copilot, and Poe). Crash risk maps and corresponding weight values were combined to produce an crash suitability map indicating the potential risk of micro-mobility crashes. Furthermore, the significance of these factors across different collision types was tested using a multinomial logistic regression model. To the best of the authors' knowledge, this is the first study to apply a macro-scale dataset and an AI-enhanced geospatial decisionmaking approach to analyse micro-mobility crashes. The findings highlight the need for local governments and urban planners to implement targeted safety measures in regions with high crash potential.
Keyword:
Traffic safety
Sustainable infrastructure
generative AI
Micro-mobility risk assessment
期刊
IF:
5.4
论文数:
655
被引数:
2.2W
机构
引用论文
Spatiotemporal comparative analysis of scooter-share and bike-share usage patterns in Washington, DC华盛顿特区滑板车共享和自行车共享使用模式的时空比较分析
OAAI
OAAI分享收藏Paving the way for autonomous Vehicles: Understanding autonomous vehicle adoption and vehicle fuel choice under user heterogeneity为自动驾驶汽车铺平道路: 了解用户异质性下的自动驾驶汽车采用和车辆燃料选择
PREAI
分享收藏Understanding factors influencing e-scooterist crash risk: A naturalistic study of rental e-scooters in an urban area了解影响电动滑板车撞车风险的因素: 对城市地区租赁电动滑板车的自然主义研究OAAI
OAAI分享收藏Gamifying the city: E-scooters and the critical tensions of playful urban mobility.png)
Mobilities
MobilitiesIF0
PREAI
分享收藏Bike share's impact on car use: Evidence from the United States, Great Britain, and Australia自行车共享对汽车使用的影响: 来自美国、英国和澳大利亚的证据OAAI
OAAI分享收藏
查看更多内容