利用XGBoost模型查明土地利用格局对行人交通事故严重程度的非线性影响  

Identifying Non-linear Effects of Land Use Patterns on Pedestrian Crash Severity Using XGBoost Decision Tree Model

作　　者：刘琪琪 陈春[1,3] 匡新晖 LIU Qi-qi;CHEN Chun;KUANG Xin-hui(Research Centre for Eco-Habitat and Green Transportation,Chongqing Jiaotong University,Chongqing 400074,China;College of Transportation,Chongqing Jiaotong University,Chongqing 400074,China;Institute of Intelligent City,Chongqing Jiaotong University,Chongqing 400074,China)

机构地区：[1]重庆交通大学,生态人居与绿色交通研究中心,重庆400074 [2]重庆交通大学交通运输学院,重庆400074 [3]重庆交通大学智慧城市学院,重庆400074

出　　处：《科学技术与工程》2025年第3期1253-1261,共9页Science Technology and Engineering

基　　金：国家自然科学基金(42071218)。

摘　　要：土地利用与交通安全是城市地理和交通运输领域共同关注的热点,但目前关于土地利用对行人交通事故的影响研究多纳入建成环境统一框架,并多采用土地利用混合度或土地利用类型占比来衡量,缺乏对土地利用类型的细化研究,难以有效指导设计实践。以重庆市渝中区为例,基于兴趣点(point of interest,POI)数据对土地利用类型进行精细刻画,应用极致梯度提升树(extreme gradient boosting,XGBoost)模型,探究土地利用类型以及行人、道路条件、道路环境等对行人交通事故严重程度影响的非线性关系。研究发现:①土地利用类型对行人交通事故严重程度有重要作用,其中影响较大的分别是医院、住宅和教育用地,事故点缓冲区300 m内存在医院、居民小区以及教育用地对行人交通事故严重程度有降低作用;②弯道和弯坡道的道路线形处是严重行人交通事故的高发区;路段进出口处、窄路等路口路段处对行人交通事故严重程度有降低作用。研究结论可为精细化的土地利用规划与治理以降低行人交通事故严重程度提供一定的政策启示。Land use and traffic safety is a hot topic of mutual concern in the fields of urban geography and transportation.However,existing research on the impact of land use on pedestrian traffic accidents often incorporates a unified framework of built environment and mainly adopts measures such as land use mix or the proportion of land use types,lacking detailed analysis on land use types,thus making it difficult to translate their findings into actionable design strategies.Taking Yuzhong District,Chongqing City as an example,the land use types were finely characterized based on point of interest(POI)data,and the extreme gradient boosting(XGBoost)model was applied to explore the nonlinear relationships between land use types,pedestrian,road conditions,road environment,and the se-verity of pedestrian traffic accidents.The study finds these as follows.①Land use types play an important role in the severity of pedes-trian traffic accidents,with hospitals,residential areas,and educational land being the most influential.The presence of hospitals,residential areas,and educational land within a 300-meter buffer zone around accident sites reduces the severity of pedestrian traffic ac-cidents.②Road sections with curved and sloped road alignments are high-risk areas for severe pedestrian traffic accidents;entrances and exits of road sections,narrow road sections,and intersections have a mitigating effect on the severity of pedestrian traffic accidents.The findings provide policy insights for refined land use planning and governance to reduce the severity of pedestrian traffic accidents.

关 键 词：土地利用 建成环境 极致梯度提升决策树(XGBoost) 交通安全 

分 类 号：U491.3[交通运输工程—交通运输规划与管理]