Incorporating polycentric development and neighborhood life-circle planning for reducing driving in Beijing: Nonlinear and threshold analysis

《Incorporating polycentric development and neighborhood life-circle planning for reducing driving in Beijing: Nonlinear and threshold analysis》

打印

关键字: Regional accessibility;Local land use;Polycentric development;Neighborhood life-circle planning;Nonlinear and threshold effects;Gradient boosting decision tree (GBDT)

作者单位: School of Urban Planning and Design, Peking University Shenzhen Graduate School, Shenzhen, Guangdong 518055, China;Key Laboratory of Earth Surface System and Human-Earth Relations of Ministry of Natural Resources of China, Peking University Shenzhen Graduate School, Shenzhen, Guangdong 518055, China;School of Urban Planning and Design, Peking University Shenzhen Graduate School, Shenzhen, Guangdong 518055, China;Key Laboratory of Earth Surface System and Human-Earth Relations of Ministry of Natural Resources of China, Peking University Shenzhen Graduate School, Shenzhen, Guangdong 518055, China

摘要: This study adopts a gradient boosting decision tree (GBDT) model to contrast the relative importance of improving regional accessibility versus local land use for reducing driving distance, as well as their nonlinear and threshold effects. The nonlinear analysis allows us to scrutinize the effective impact scopes of urban and suburban centers, which can provide planning guidelines for polycentric development, while threshold effects can help planners determine normative ranges of land use settings for planning pedestrian-scale neighborhoods (e.g., the 15-minute life-circle plan in China). Based on a 2017 travel diary survey in Beijing, empirical findings reveal the synergistic effects of the proximities to urban and suburban centers on driving distance, demonstrating the importance of polycentric development for reduced driving. In addition, many local land use variables have distinctive threshold effects, including local access to daily facilities and density indicators. For driving reduction, effective densities of health, civil, and retail/service facilities are around 15, 10, and 200 establishments (points of interest) per km2, respectively. These findings demonstrate the importance of nonlinear and threshold analysis and suggest incorporating polycentric development and neighborhood life-circle planning would lower vehicle use in Beijing.