To explore the spatiotemporal interaction between the built environment and taxi travel demand, this study utilizes taxi order data from Kunming City, constructing built environment indicators from two dimensions: land use and transportation infrastructure. By applying Multiscale Geographically Weighted Regression (MGWR) and Geographically and Temporally Weighted Regression (GTWR) models to the origin-destination (OD) points, the spatial heterogeneity and temporal non-stationarity of built environment variables and their impacts on taxi demand were examined. The results indicate that major transportation hubs, metro stations, dining facilities, and parking lots are key factors influencing taxi demand, exhibiting significant spatial heterogeneity across OD points. In contrast, cultural-educational and medical institutions show no spatial heterogeneity in taxi demand. Notably, commercial enterprises, recreational facilities, and parks display markedly different spatial heterogeneity between OD points. Further analysis reveals that built environment variables also exhibit significant temporal heterogeneity, generally following two trends: an initial decline followed by a rise, and an initial rise followed by a decline. Detailed spatiotemporal analyses of peak hours reveal that major transportation hubs significantly impact demand at destination points during morning peaks, while their effect on origin points is more pronounced during evening peaks. Parking lots promote demand at both origin and destination points, peaking during evening rush hours, with a stronger influence on destination points. Lastly, the study uncovers that bus stops cooperate with taxi demand during morning peaks but exhibit a competitive relationship during evening peaks.
然而,这些研究通常未能充分考虑起讫点空间流数据的时空尺度敏感性和异质性问题。为此,部分学者尝试通过引入时空回归模型解决这一问题。Wang等[14]引入地理加权回归(Geographically weighted regression,GWR)模型,研究了一天中不同高峰时段对网约车出行需求的重要影响因素,发现建成环境变量在地理空间上存在异质特性,与普通最小二乘法(Ordinary least squares,OLS)模型相比,GWR模型能提供关于建成环境变量影响的空间变化详细信息。但是,GWR模型忽略了异质建成环境变量之间存在空间尺度差异的问题,导致模型结果不够完善,且忽视了起讫点空间流数据的时空尺度敏感性和异质性。为此,部分学者使用多尺度地理加权回归(Multiscale geographically weighted regression,MGWR)模型和半参数地理加权回归模型[1]进行更为完善的空间异质性分析,如李想等[15]使用MGWR模型探究了建成环境对地铁-公交客流的影响及尺度效应,该模型能够进行不同尺度下的空间异质分析,但仍未将时间因素真正纳入,以揭示时间异质性。
综上,考虑起讫点建成环境异质性有助于区分起始点与到达点建成环境对出租车出行需求影响程度的差异,这对于进一步提升出租车服务质量至关重要。同时,在明确起讫点范围内建成环境和出租车出行需求影响存在差异性的基础上,需充分考虑建成环境变量异质性的差异,同时将时间因素纳入GWR模型,因此本文充分利用MGWR模型挖掘空间异质性,利用GTWR模型呈现时间异质性,最后,基于出租车订单数据、兴趣点(Point of interest,POI)数据及路网数据,利用GWR改进的MGWR模型和GTWR模型,从土地利用和交通基础设施两个维度,探究出租车出发点和到达点周围建成环境对出租车出行需求的影响及时空异质效应。
建成环境数据从土地利用和交通基础设施两个维度进行构建,基于高德开发平台,获取2021年期间POI数据,土地利用数据中包括餐饮、购物场所、公园景点、公司企业、科教文化、体育休闲、医疗机构、政府机构等类型,交通基础设施数据中包括地铁站点、公交站点、停车场、机场、高铁站、汽车站、渡口等设施数据,以此将地理位置数据累积到研究单元,计算出相应的建成环境指标。路网数据(2021)来自Open Street Map,用于计算路网密度。
MaJian-xiao, ZhaoFei-yan, YinChao-ying, et al. Spatial-temporal heterogeneity effects of built environment and taxi demand on ride-hailing demand[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(5): 136-145.
[3]
GhaffarA, MitraS, HylandM. Modeling determinants of ride sourcing usage: a census tract-level analysis of Chicago[J]. Transportation Research Part C: Emerging Technologies, 2020, 119: No.102769.
ZhuangHao-ming, LiuXiao-ping. Identifying changes in urban spatial structure using taxi trajectory data: a case study in Shenzhen[J]. Tropical Geography, 2020, 40(2): 217-228.
ZhengXiao-lin, LiuQi-liang, LiWen-kai, et al. Discovery of urban human mobility patterns from smart card transactions and taxi GPS trajectories: a comparative study[J]. Journal of Geo-information Science, 2020, 22(6): 1268-1281.
JiXiao-feng, DengRuo-fan, QiaoXin, et al. Nonlinear influence of built environment on temporal aggregation modes of shared bicycles[J]. Journal of Jilin University (Engineering and Technology Edition), 2025, 55(7): 2233-2242.
ChenQi-xiang, LvBin, ChenXi-qun, et al. Impacts of built environment on competition and cooperation relationship between taxi and subway considering spatial hetero geneity[J]. Journal of Transportation Systems Engineer ing and Information Technology, 2022, 22(3): 25-35.
PengShi-yao, ChenShao-kuan, XuQi, et al. Spatial characteristics of land use based on POI and urban rail transit passenger flow[J]. Acta Geographica Sinica, 2021, 76(2): 459-470.
[14]
EwingR, CerveroR. Travel and the built environment: a synthesis[J]. Transportation Research Record, 2001, 1780(1): 87-114.
GuanHao-tian, JiXiao-feng, LiWu, et al. Influence of built environment on integrated use of bike sharing and metro[J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 24(4): 200-211.
PanYi-yong, XuJia-cong, YouYi-wen, et al. Multi-scale spatial heterogeneity analysis of influencing factors of ride-hailing travel demand[J]. Journal of Jilin University (Engineering and Technology Edition), 2025, 55(5): 1567-1575.
RenPeng, PengJian-dong, YangHong, et al. Relationship between jobs-housing balance and built environment in areas around urban rail transit stations of Wuhan[J]. Journal of Geo-information Science, 2021, 23(7): 1231-1245.
LiXin-ran, HeRi-xing, JiangChao, et al. A multiscale spatial prediction model for taxi OD flow based on deep gravity and its interpretability research in Beijing[J]. Journal of Geo-information Science, 2024, 26(6): 1390-1406.
LiQi-lin, LuYing-rong, DingChuan, et al. Analysis of the impact of built environment on taxi travel demand considering spatial spillover effect[J]. Journal of Transportation Systems Engineering and Information Technology, 2016, 16(5): 39-44.
[25]
WangS, NolandR B. Variation in ride-hailing trips in Chengdu, China[J]. Transportation Research Part D: Transport and Environment, 2021, 90: No.102596.
LiXiang, YanQi-peng, LuoChen. Impact of built environment on flow of transfer passengers between subway and bus considering spatial heterogeneity[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(2): 100-110.
WangJian-yu, ChenXian-tian, JiaoPeng-peng, et al. Interactive effect on traffic accident severity considering built environment[J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 24(2): 272-280.
XianHao-ran, JiangHao, LiaoJuan, et al. Study on taxi trajectory prediction method based on transformer[J]. Engineering Journal of Wuhan University, 2025, 58(2): 306-315.
MaJian-xiao, YinJie-xiang, ShaoChun-fu, et al. The spatiotemporal heterogeneity of the impact of the built environment on taxi travel demand[J], Journal of Beijing Jiaotong University, 2025, 49(1): 119-127.
LiCong-ying, WuJia-xi, ZhangHong-tao, et al. Spatio-temporal heterogeneous effects of built environment on urban motorized travel[J]. Journal of Transportation Engineering and Information, 2024, 22(4): 52-66.
MaJun-ze, ZhengChang-jiang, WuFei, et al. Analysis of characteristics and influencing factors of dockless bike sharing connected to metro[J]. Journal of Jilin University (Engineering and Technology Edition), 2025, 55(8): 2639-2650.
LongXue-qin, ZhaoHuan, ZhouMeng, et al. Spatiotemporal heterogeneity of the impact of built environment in Chengdu on online car-hailing passengers' pick-up points[J]. Scientia Geographica Sinica, 2022, 42(12): 2076-2084.
[40]
VandenbulckeG, DujardinC, ThomasI, et al. Cycle commuting in Belgium: spatial determinants and 're-cycling' strategies[J]. Transportation Research Part A: Policy and Practice, 2011, 45(2): 118-137.
ChengXiao-yun, ZhangXue-yu, ShiCheng, et al. Analysis of spatial effect and its heterogeneity on night-time travel based on multi source data[J]. China Journal of Highway and Transport, 2021, 34(12):288-301.