基于可塑性面积单元问题的建成环境对交通状态的影响

The impact of built environment on traffic conditions based on modifiable areal unit problem

针对分析建成环境对交通状态影响时网格单元划分问题,以济南市为研究区域,选取人口密度、购物商场密度、公交站点密度作为济南市建成环境的评价指标,选取车辆轨迹点平均速度作为济南市交通状态的评价指标,考虑可塑性面积单元问题(modifiable areal unit problem,MAUP),分别采用1 km×1 km、5 km×5 km、10 km×10 km的3种正方形网格及1 km×5 km、1 km×10 km、5 km×10 km的3种长方形网格划分研究区域,通过皮尔逊相关系数、普通最小二乘法(ordinary least squares,OLS)、空间自相关分析3种方法分析不同划区方案下建成环境与交通状态间的相关性,通过计算并比较建成环境与交通状态间的皮尔逊相关系数、OLS拟合优度、莫兰指数确定最佳划区方案。结果表明:以5 km×5 km的正方形网格单元划分研究区域时,交通状态和建成环境间的相关性最大,可较好地反映建成环境对交通状态的影响。

Aiming at the problem of grid unit division when analyzing the impact of the built environment on traffic conditions.Using Jinan City as the study area,population density,shopping mall density,and bus stop density are selected as evaluation indicators for the built environment,the average speed of vehicle trajectory points is selected as the evaluation indicator for traffic conditions.Considering the modifiable areal unit problem(MAUP),three square grids of 1 km×1 km,5 km×5 km,10 km×10 km,and three rectangular grids of 1 km×5 km,1 km×10 km,5 km×10 km are used to divide the study area.The correlation between traffic conditions and the built environment is analyzed using Pearson correlation coefficient,ordinary least squares(OLS),and spatial autocorrelation analysis.The best grid division scheme is determined by comparing Pearson correlation coefficients,OLS goodness of fit,and Moran′s I.The results showed that when a 5 km×5 km square grid is used as the grid unit to divide the study area,the correlation between traffic congestion and built environment is the greatest,which can better reflect the impact of built environment on traffic status.