A High-Resolution, Yet Scalable Framework for Transport Infrastructure Accessibility Based on Open Big Data

Throughout the last years, open big data has brought many challenges and revealed much potential for transportation research. In this study, we show how large open datasets can be used to accurately estimate the accessibility of transportation infrastructure for whole China, a scale unprecedented in the literature. Hundreds of airports and almost one thousand high-speed railway (HSR) stations serve the tremendously growing transportation demand. These access points to the transportation networks are not equivalently distributed throughout the country. Existing large-scale studies on transportation infrastructure accessibility focus on highly-aggregated forms of land-use area, with travel times from centroids of administrative regions. Other studies analyze spatially-constrained regions with very detailed data. In this study, we design and implement a fine-grained accessibility framework based on large open data sets, which allows us to estimate the accessibility based on grid cells with a resolution of less than one square kilometer. We automatize the computation of free-flow road travel times, as well as, the public transit times, from grid cells to infrastructure elements, by using a scalable routing framework. Based on our experiments, we find that the HSR network for China is much better accessible for the population than airports. Moreover, grid cells with a high population density are much better connected to the Chinese transportation networks than other grid cells. Our methodology is generic in that it can be applied on an even larger scale (whole planet) and with different points of interest.