Mobility Signatures: A Tool for Characterizing Cities Using Intercity Mobility Flows.

Understanding the patterns of human mobility between cities has various applications from transport engineering to spatial modeling of the spreading of contagious diseases. We adopt a city-centric, data-driven perspective to quantify such patterns and introduce the mobility signature as a tool for understanding how a city (or a region) is embedded in the wider mobility network. We demonstrate the potential of the mobility signature approach through two applications that build on mobile-phone-based data from Finland. First, we use mobility signatures to show that the well-known radiation model is more accurate for mobility flows associated with larger Finnish cities, while the traditional gravity model appears a better fit for less populated areas. Second, we illustrate how the SARS-CoV-2 pandemic disrupted the mobility patterns in Finland in the spring of 2020. These two cases demonstrate the ability of the mobility signatures to quickly capture features of mobility flows that are harder to extract using more traditional methods.

Predicting subway passenger flows under different traffic conditions.

Passenger flow prediction is important for the operation, management, efficiency, and reliability of urban rail transit (subway) system. Here, we employ the large-scale subway smartcard data of Shenzhen, a major city of China, to predict dynamical passenger flows in the subway network. Four classical predictive models: historical average model, multilayer perceptron neural network model, support vector regression model, and gradient boosted regression trees model, were analyzed. Ordinary and anomalous traffic conditions were identified for each subway station by using the density-based spatial clustering of applications with noise (DBSCAN) algorithm. The prediction accuracy of each predictive model was analyzed under ordinary and anomalous traffic conditions to explore the high-performance condition (ordinary traffic condition or anomalous traffic condition) of different predictive models. In addition, we studied how long in advance that passenger flows can be accurately predicted by each predictive model. Our finding highlights the importance of selecting proper models to improve the accuracy of passenger flow prediction, and that inherent patterns of passenger flows are more prominently influencing the accuracy of prediction.