公交出行时空模式的影响因素和效应研究

Exploring the factors influencing the spatio-temporal patterns of bus travel and their effects

为揭示公交出行时空模式的影响因素和作用效应,利用智能公交数据获取公交出行时空特征,应用DBSCAN聚类算法挖掘公交出行的时空规律性。以起点建成环境、终点建成环境、出行路径性能、公交运营性能和公交出行强度5个方面的18个指标为自变量,以公交出行时空特征和公交出行时空规律性2个方面的4个指标为因变量构建变量指标体系,融合智能公交、POI等多源数据对指标进行量化计算,建立结构方程模型对公交出行时空模式的影响效应进行分析。研究结果表明:公交出行模式具有时空异质性,平峰时段公交出行距离和出行时长均长于早、晚高峰时段;时间规律性强的公交出行主要发生在高峰时段,空间规律性强的公交出行集中分布在城市中心区;出行路径性能、公交运营性能、起点建成环境、终点建成环境对公交出行的时空特征有显著影响(早高峰影响效应分别为-0.749,-0.413,-0.244,-0.228);公交出行强度、出行路径性能、起点建成环境和终点建成环境对公交出行的时空规律性有显著影响(早高峰影响效应分别为0.688,0.069,0.022,0.021);各因素对公交出行时空模式的影响具有时间异质性,平峰时段的影响均大于早、晚高峰时段。研究结论能够为公交导向的城市规划设计、公交系统优化等工作提供参考依据。

To better understand the factors influencing the spatio-temporal patterns of bus travel and their effects,data from the Advanced Public Transportation System(APTS)were used to obtain the spatio-temporal characteristics of bus travel,and the Density-Based Spatial Clustering of Applica-tions with Noise(DBSCAN)clustering algorithm was applied to calculate its spatio-temporal regu-larity.18 indices of 5 aspects(origin building environment,destination building environment,travel route performance,bus operation performance and bus travel intensity)as independent variables and 4 indices of 2 aspects(spatio-temporal characteristics and spatio-temporal regularity of bus travel)as dependent variables were used to construct the system of variables.The indices were quantified by integrating multi-source data including APTS,POI etc.The structural equation model was estab-lished to analyze the effects on the spatio-temporal patterns of bus travel.The results show that bus travel patterns have spatio-temporal heterogeneity,as the distance and time of bus travel during non-peak hours are longer than those during the morning and evening peaks.The bus trips with strong temporal regularity mainly occur in the peak hours,while the bus trips with strong spatial regularity are concentrated in the city center.Travel route performance,bus operation performance,origin building environment,and destination building environment have significant effects on the spatio-temporal characteristics of bus travel(-0.749,-0.413,-0.244,and-0.228,respectively,at the morning peak).The bus travel intensity,travel route performance,origin building environment,and destination building environment all have significant effects on the spatial-temporal regularity of bus travel(0.688,0.069,0.022,and 0.021,respectively,at the morning peak).The influence of all factors on the spatio-temporal patterns of bus travel have time heterogeneity,and the influence during non-peak hours is greater than that during the morning and evening peaks.These conclusions can provide the basis for a reference for public transportation-oriented urban planning and system optimization.