Identifying Key Nodes in Urban Transportation Systems Using the Information Diffusion Model

Urban transporting networks are a type of complex systems, widely existing in modern society. The identification of nodes with significant influential ability is crucial for scientific researches and practical applications including urban planning and city management. Through the survey of relevant literature, most of existing methods for determining influential nodes in cities ignore the topological performance or only consider an agent-based approach using structural metrics for nodal detections, which requires multi-dimensional data, resulting in insufficient robustness. To address such shortcomings, this paper first designs a metric to assess the influential ability of nodes via their structural information. Equipped with the metric, an evolution algorithm is implemented to detect a set of nodes with distinct influential ability. In the experiment, a road network data processor using Pipe4data is developed to process urban road networks from Haizhu District, Guangzhou. The effectiveness of the proposed method has been experimentally verified.