Road Network Accessibility Analysis Based on Dynamic Diversity Entropy

In this work, we propose a novel accessibility measure based on Dynamic Diversity Entropy to evaluate the difficulty of citizens accessing various nodes with resources based on network structure. Furthermore, the self-avoiding random walk is adopted for improving the computational efficiency of evaluating accessibility of nodes in a scaled urban street network. Dynamic Diversity Entropy is introduced to measure the outcome of self-avoiding random walks. Then, we introduce link travel times representing the time-varying traffic status, and spatial heterogeneous Point of Interest’s density representing nodes’ resource weights in the accessibility assessment is used. A case study on a partial road network inside Beijing’s third ring road is conducted. The spatial accessibility distribution and the relationship between accessibility to restaurants and time limit are visualized to investigate the performance of the proposed accessibility assessment indicators.