为更好地调度出租车运力，缓解热点载客区域出租车供需不平衡现象，需探究出租车需求的时空分布特征及其影响因素。鉴于此，基于出租车GPS数据、计价器数据、公共交通刷卡数据和兴趣点（Point of Interesting，POI）数据等多源异构数据，结合相关性分析法对区域出租车出行需求影响因素进行筛选，建立多维度的影响因素集，构建基于地理加权回归的区域出租车出行需求影响模型。以北京市1 398个交通小区的数据为例，分析不同时空条件下各影响因素对出租车出行需求的影响程度。结果表明：出租车出行需求空间分布具有空间集聚效应，影响因素对出租车需求的影响程度具有空间非稳态特征；各中心区域住宅密度、周边且公司密集区域办公密度及城市外围区域的休闲娱乐服务密度对出租车出行需求有很强的正影响；城市外围区域住宅密度、各中心区域办公密度与出租车出行需求呈负相关；非工作日休闲娱乐服务密度对出租车出行需求促进作用明显大于工作日；区域公共交通产生量对出租车出行需求的影响早、晚高峰差异显著。通过模型对比分析可知，所建模型具有较高的精度，适用于解释各影响因素对出租车出行需求影响的时空差异性。
In order to better dispatch taxis′ transportation capacity so as to alleviate the imbalance between taxi supply and demand in high passenger demand areas, it is important to accurately explore the spatiotemporal distribution characteristics of taxi travel demand and its influencing factors. Based on the multi-source heterogeneous data including taxi GPS data, taximeter data, public transportation transactions data and Point of Interesting(POI) data, correlation analysis methods were used to screen the influencing factors of taxi travel demand and a multi-dimensional factor set was established. Then the influencing model was put forward based on Geographical Weighted Regression(GWR). Taking the data of 1 398 traffic districts in Beijing as an example, the impact of various factors on taxi travel demand under different space-time conditions was explored. The results show that the spatial distribution of taxi travel demand has spatial agglomeration effect, and the influence degree of the variables has spatial nonstationary characteristic; the residential density in the central areas, the office density in the surrounding and company-intensive areas and the entertainment service density in the peripheral areas of the city have a strong positive impact on taxi travel demand; the residential land density in the peripheral areas and the office density in the central areas are negatively correlated with taxi travel demand; the promotion effect of entertainment service density on taxi travel demand in nonworking days is significantly greater than that in working days; the impact of regional public transportation generation volume on taxi travel demand shows significant difference between early peak and late peak. The comparative analysis of the models shows that the proposed model has high accuracy and is suitable for explaining the spatial and temporal differences of the influencing factors on taxi travel demand.
[1] 王芮. 基于GPS数据的城市出租车出行需求研究[D]. 济南：山东大学，2016.
[2] Ingvardson J B, Nielsen O A. Effects of New Bus and Rail Rapid Transit Systems: an International Review[J]. Transport Reviews, 2018, 38(1): 96-116.
[3] Zhang J, Shen D, Tu L, et al. A Real-Time Passenger Flow Estimation and Prediction Method for Urban Bus Transit Systems[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(11): 3168-3178.
[4] Taylor B D, Fink C N Y. The Factors Influencing Transit Ridership: A Review and Analysis of the Ridership Literature[R]. Berkeley CA: University of California Transportation Center, 2003.
[5] Gutiérrez J, Cardozo O D, García-Palomares J C. Transit Ridership Forecasting at Station Level: An Approach Based on Distance-Decay Weighted Regression[J]. Journal of Transport Geography, 2011, 19(6): 1081-1092.
[6] Ding C, Chen P, Jiao J F. Non-linear Effects of the Built Environment on Automobile-Involved Pedestrian Crash Frequency: A Machine Learning Approach[J]. Accident Analysis and Prevention, 2018, 112: 116-126.
[7] 姜伟. 动态因素影响下居民租车出行选择模型构建与应用研究[D]. 重庆：重庆交通大学，2017.
[8] 李龙. 成都市居民出行方式选择影响因素研究[D]. 成都：西南交通大学，2016.
[9] Kuby M, Barranda A, Upchurch C. Factors Influencing Light-rail Station Boardings in the United States[J]. Transportation Research Part A: Policy and Practice, 2004, 38(3): 223-247.
[10] 贾佃通. 快速公交客流影响因素分析及宏观预测模型研究[D]. 哈尔滨：哈尔滨工业大学，2014.
[11] Bradley W L, Kobayashi T. Spatial Heterogeneity and Transit Use[R]. Bloomington: Department of Geography, Indiana University, 2007.
[12] Cardozo O D, García-Palomares J C, Gutiérrez J. Application of Geographically Weighted Regression to the Direct Forecasting of Transit Ridership at Station-level[J]. Applied Geography, 2012, 34: 548-558.
[13] Qian X W, Ukkusuri S V. Spatial Variation of the Urban Taxi Ridership Using GPS Data[J]. Applied Geography, 2015, 59: 31-42.
[14] 张俊杰. 基于地理加权回归模型公交客流影响因素的空间异质性研究[D]. 西安：长安大学，2018.
[15] Sung H, Oh J T. Transit-Oriented Development in a High-Density City: Identifying Its Association with Transit Ridership in Seoul, Korea[J]. Cities, 2011, 28(1): 70-82.
[16] 李旭，仇蕾洁，姜鑫. 我国31省份城乡人均医疗保健支出空间自相关差异性研究[J]. 中国卫生经济，2019，38（1）：42-46.
[17] Fotheringham A S, Charlton M, Brunsdon C. Measuring Spatial Variations in Relationships with Geographically Weighted Regression[M]. Berlin: Springer-Verlag Berlin Heidelberg, 1997: 60-80.
京公网安备 11010102003417号
京ICP备17006742号-3