Urban Environment-Oriented Traffic Zoning Based on Spatial Cluster Analysis

Abstract

Environment-oriented traffic management zoning (TMZ) provides an important way to study spatial variation of urban traffic environment. The TMZ can be used to manage the urban traffic environment and establish regional improvement strategies. In this study, a spatial immune cluster method based on aiNet (Artificial Immune Network) is designed to facilitate the traffic environment zoning. Spatial coordinates are employed to measure the spatial correlation of the traffic system, and the weightings of different indices are calculated with the use of PCA (principle component analysis). For the case of Beijing, the whole city area is divided into evenly distributed grid cells. An urban traffic environment index system is developed to measure the traffic environment condition of different grids based on the concept of traffic sustainability. The GIS (Geographical information System) software of ArcGIS and database programming are used to calculate the indexical values of each grid. In order to classify the grid data set, we use two analysis methods, i.e. immune cluster algorithm and K-mean algorithm, and it is found that the immune cluster algorithm has better performance for the classification. With the use of this method, the study area is divided into 4 zones.

Keywords: traffic environment, management zones, spatial clustering, geographical information system, principal component analysis, K-mean, immune clustering