| 0 | 0 | 1 |
| 下载次数 | 被引频次 | 阅读次数 |
随着城市化进程的不断推进，精细尺度的城市职住人口空间分布制图研究逐渐成为城市规划和管理中的关键课题。本文以上海市为研究区，基于上海市兴趣点（POI）数据，提出了一种基于随机森林算法的职住人口空间分布模拟模型，用于绘制1 km尺度的上海市职住人口空间分布。首先，提取多类POI的空间分布特征，结合手机信令数据构建职住人口分布训练样本集；其次，利用随机森林模型拟合人口空间化关系，并基于变量重要性识别职住空间分异的关键驱动因子；最后，结合人口普查数据完成结果校正。结果表明，本文所提议模型拟合的平均绝对误差为155～235人，决定系数达0.74～0.82，整体精度良好；居住人口主要受生活服务和餐饮服务设施驱动，就业人口受公共交通和公司企业设施显著影响，职住混合功能区则受多类设施综合作用；上海市职住人口呈现“核心集聚—副中心支撑—外围分散”的分层空间格局。本文研究成果可为城市职住空间优化、通勤效率提升及规划决策提供数据支撑与科学依据。
Abstract：With the continuous growth of urbanization, simulating the jobs-housing spatial distribution at a fine scale has gradually become a key issue in urban planning and management. Hence, this paper proposes an jobs-housing spatial distribution simulation model based on points of interests(POIs) data using the random forest algorithm by taking Shanghai City as the study area. Firstly, spatial distribution features of different categories of POI in Shanghai City are extracted, and a training sample set for jobs-housing population distribution is constructed by integrating mobile phone signaling data.Secondly, the random forest model is employed to fit the relationship underlying population spatialization,and the key driving factors of jobs-housing spatial differentiation are identified based on variable importance analysis. Finally, the proposed simulated results are calibrated using census data. The results show that the proposed model can get a mean absolute error of 155 to 235 persons per grid and coefficient of determination values of 0.74 to 0.82 on the testing data, demonstrating high predictive accuracy. Residential population distribution is mainly driven by living service and catering facilities, whereas employments distribution is significantly influenced by public transport and company-related facilities. Mixed jobshousing functional areas are jointly affected by multiple categories of facilities. The spatial distribution of employing and housing populations in Shanghai City exhibits a hierarchical pattern characterized by core agglomeration, subcenter support, and peripheral dispersion. The findings from this paper can provide data support and a scientific basis for optimizing urban jobs-housing space, improving commuting efficiency, and making informed planning decisions.
[1]Cong Zhu,Fang Chuanglin,Zhang Lifang,et al.Simulating the interrelationships among population, water, ecology,and economy in urban agglomerations based on a system dynamics approach[J]. Journal of cleaner production,2024,439140813.
[2]Haimeng Liu,Cui Weijia,Zhang Mi. Exploring the causal relationship between urbanization and air pollution:Evidence from China[J]. Sustainable cities and society,2022,80103783.
[3]刁晶晶，王倩，程苑，等.职住平衡对居民出行能耗的影响研究回顾[J].交通工程，2019,19(5):25-30.
[4]宋金平，刘庆芳，张学波，等.职住空间关系演化过程与展望[J].人文地理，2024,39(6):33-39.
[5]牟凤云，朱智贤，汪孝之，等.基于建成环境的超大城市职住人口空间分布模拟：以重庆市中心城区为例[J].科学技术与工程，2024,24(35):14930-14939.
[6]吴晓虎，王贺封，赵傲，等.基于多源数据与XGBoost模型的上海市人口空间化研究[J].测绘与空间地理信息，2024,47(9):33-36.
[7]董南，杨小唤，蔡红艳.人口数据空间化研究进展[J].地球信息科学学报，2016,18(10):1295-1304.
[8]吴华意，胡秋实，李锐，等.城市人口时空分布估计研究进展[J].测绘学报，2022,51(9):1827-1847.
[9]Rong Wu,Wang Jieyu,Zhang Dachuan,et al. Identifying different types of urban land use dynamics using Point-ofinterest(POI)and Random Forest algorithm:The case of Huizhou, China[J].Cities,2021,114103202.
[10]关庆锋，任书良，姚尧，等.耦合手机信令数据和房价数据的城市不同经济水平人群行为活动模式研究[J].地球信息科学学报，2020,22(1):100-112.
[11]吕安民，李成名，林宗坚，等.人口统计数据的空间分布化研究[J].武汉大学学报（信息科学版），2002,27(3):301-305.
[12]闫庆武，卞正富，张萍，等.基于居民点密度的人口密度空间化[J].地理与地理信息科学，2011,27(5):95-98.
[13]Dejan Grigillo,Kosmatin Fras Mojca,Petrovi V C Du-V-SAn. Automated building extraction from IKONOS images in suburban areas[J]. International Journal of Remote Sensing,2012,33(16):5149-5170.
[14]Francisco-Javier Gallego,Batista Filipe,Rocha Carla,et al.Disaggregating population density of the European Union with CORINE land cover[J].International Journal of Geographical Information Science,2011,25(12):2051-2069.
[15]谭敏，刘凯，柳林，等.基于随机森林模型的珠江三角洲30 m格网人口空间化[J].地理科学进展，2017,36(10):1304-1312.
[16]康停军，张新长，赵元，等.基于GIS和多智能体的城市人口分布模拟[J].中山大学学报（自然科学版），2012,51(3):135-142.
[17]Zhifeng Cheng,Wang Jianghao,Ge Yong. Mapping monthly population distribution and variation at 1-km resolution across China[J]. International Journal of Geographical Information Science,2022,36(6):1166-1184.
[18]Ryohei Kato,Luo Lichen,Fourie Pieter-Jacobus,et al.Evaluating and Validating Stay Point Detection Algorithms with different GPS Log Intervals[J]. Procedia computer science,2024,23(8):803-808.
[19]Jing Yuan,Zheng Yu,Xie Xing. Discovering regions of different functions in a city using human mobility and POIs[A]//Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining[C]. New York, NY, USA:ACM, 2012:186-194.
[20]Simon-J Sheather. Density estimation[J].Statistical science,2004,588-597.
[21]L Breiman. Random forest[J]. Machine Learning,2001,455-32.
[22]Yao Yao,Qian Chen,Hong Ye,et al. Delineating Mixed Urban“Jobs-Housing”Patterns at a Fine Scale by Using High Spatial Resolution Remote-Sensing Imagery[J].Complexity,2020,2020(1):8018629.
[23]Ulrike Gr O Mping. Variable importance assessment in regression:linear regression versus random forest[J].The American Statistician,2009,63(4):308-319.
[24]Pierre Deville,Linard Catherine,Martin Samuel,et al.Dynamic population mapping using mobile phone data[J]. Proceedings of the National Academy of Sciences,2014,111(45):15888-15893.
基本信息:
DOI：10.20006/j.cnki.61-1363/P.2026.02.003
中图分类号:C922;P208;TU984.2
引用信息:
[1]党威龙.基于POI数据的职住人口空间分布模拟研究——以上海市为例[J].测绘技术装备,2026(02):22-30.DOI:10.20006/j.cnki.61-1363/P.2026.02.003.
2026-07-14
2026-07-14
2026-07-14