Common network characteristics of four bus transport networks in Northeast China based on a perfect space P

Abstract

The space P and space L representation models are often adopted together in the empirical analysis of bus transport networks (BTNs). In this paper, we develop a new representation model for BTNs using double edge weights and node weights, namely perfect space P. The model incorporates the number and lengths of directed routes between the two stations and the number of routes that connect to each station. Based on the model, we develop an empirical study on four large BTNs in Northeast China. Some common network characteristics of the BTNs are revealed and discussed in detail. The results show that several empirical distributions follow an exponential law. Moreover, the unweighted path length distributions can be fitted by a Gaussian function, while the weighted path length distributions can be fitted by a composite Gaussian function. Specifically, we introduce the weighted-degree distribution with different lengths and the special edge weight distribution, given the new evidence for the random behavior during the expansion of BTNs, including new stations and routes.