2020 Volume 75 Issue 6 Pages I_199-I_205
Many researchers have analysed road networks by considering their components such as node, link and connection. On the other hand, only few researches refer to tangible factors of road network such as length and orientation of each road. In this paper, we propose the new evaluation method of road network by using directional statistics. At first, we draw the graph of orientation of each road around a station. Next, we develop the method to estimate parameters of probability density function of the orientation. Then, we employ the clustering method on this estimated parameters to classify road networks around stations. Finally, we showed the applicability of the proposed method by applying them to over 1,500 stations in Tokyo Metropolitan Area.
