基于SAR模型的交通流数据分析——以贵阳市交通数据为例

Analysis of Traffic Flow Data Based on SAR Model——A Case Sudy of Guiyang City

道路的空间关联性对交通流量有重要影响。为提高交通流量预测的准确性,运用空间分析方法,基于贵阳市中心城区交通流量等数据,构建不同的空间权重矩阵,通过Moran I指数判断交通流量的空间自相关性,运用核密度估计法对不同时段交通流量的空间分布趋势面进行研究,通过空间自回归模型(Spatial auto-regression model SAR)将平均车速、车道占有率对交通流量的贡献率进行分析,并与经典线性回归模型进行比较。此研究有助于探索城市路网中交通流的热点和冷点区域,了解整个城市交通系统的运行状态,并对动态交通管理和长期城市交通规划提供决策支持。

The spatial correlation of the road has an important influence on the traffic flow. In order to improve the accuracy of traffic flow prediction, this paper uses spatial analysis method based on the data of traffic flow in thecity center of Guiyang. According to constructing different spatial weight matrix and Moran I index, we analysis the spatial autocorrelation of traffic flow and explore the spatial distribution of traffic flow in different periods using thekernel density estimation method. The contribution rate of the average vehicle speed and lane occupancy rate totraffic flow is analyzed by the spatial auto regression model, and the results are compared with the results of the classical linear regression model. This research method can be used to explore the hot spot and cold spot area of traffic flow in urban road network, and understand the whole operation state of urban transportation system, which will provide the significant support to dynamic traffic management and urban transportation plan in the long term.