Research on Evaluation Model of Road Congestion of Tourism Scenic Spots Based on Spatial Syntax and Artificial Neural Network Method—A Case of Kulangsu Island, Xiamen, China

Abstract

In today's tourism development, land-use and transportation planning are the two tasks that tend to be separated, which in turn leads to unreasonable application of roads. In order to accurately predict the flow of passengers, this paper uses Space Syntax and Artificial Neural Network (ANN) to construct the evaluation model of tourism road congestion. The model makes use of the characteristics of objective and dynamic assignment of ANN to train and estimate each scenic spot weight. The accessibility analysis of Space Syntax is used to understand the connections between roads; then it uses the mathematical model to effectively combine them, so that the model has the ability to estimate road congestion when the street network structure, scenic spots distribution and crowd flows are inconsistent. The empirical results show that: (I) The areas with high scenic spots attractiveness of Kulangsu are mainly distributed on the edge of the island; (II) The model can predict the problem of unreasonable distribution of roads. The results can be used as the basis for future tourism space management.