Fractal dimension of job-housing flows: A comparison between Beijing and Shenzhen

Abstract Job-housing flows are among the most important travel flows in a city. Therefore, it is essential to measure the extent to which the job-housing flow distribution covers the city space, which can be used as evidence for evaluating the land use status and transportation system. Existing studies have analyzed the spatial distribution of job-housing flows in terms of where and how the flows are aggregated in the city, but they have not revealed the extent to which the flow distribution fills the city space. In this study, we introduce a fractal dimension to measure the space-filling degree of the job-housing flows, which is defined based on the flow space, with the flow being the basic element. Because the flow fractal dimension is independent of the observation scale, we compared the box-counting dimensions of job-housing flows in Beijing and Shenzhen using mobile phone data. The results demonstrated that the fractal dimension was substantially higher in Beijing than in Shenzhen, indicating that the job and home distributions fill more space in Beijing, and the links between the job and home locations are more disordered and irregular in Beijing. These results are related to the more crammed urban land use and higher commuting demand from the suburbs to the city center.