Search

The integration of multi-source and diverse spatio-temporal travel data provides a comprehensive insight into urban mobility. Using data from Shenzhen's public transportation system, this study presents an analytical framework based on multiplex networks to examine variations in multi-mode public transportation usage (metro, bus, taxi, and shared bike) and their correlation with the built environment. This framework encompasses the analysis of network topological characteristics, centrality, and communities. The examination of network topological characteristics reveals that the multiplex transportation network exhibits high global accessibility and local connectivity. Network centrality analysis, focusing on weighted outdegree centrality, captures the patterns of public transportation ridership. Centrality modeling, employing the light gradient boosting machine, demonstrates a nonlinear relationship between ridership and the built environment. Factors including population density, residential land use percentage, entertainment service density, restaurant density, and metro station density consistently exhibit positive correlations with ridership across different times of the day. The community structure analysis, using consensus community detection, indicates that distinct urban areas exhibit clustering behavior based on public transportation demand patterns, forming distinct communities that closely align with the functional zoning of urban planning. These findings could provide valuable insights for the strategic planning of transportation services and the built environment. • Multi-source public transportation data are integrated to investigate urban mobility. • A multiplex network-based framework is developed to explore public transportation use. • Topological structure analysis reveals high global accessibility and local connectivity. • Centrality analysis reveals nonlinear correlations between ridership and the built environment. • Communities within multiplex networks closely align with the urban functional zoning. [ABSTRACT FROM AUTHOR]

The free-floating bike (FFB) program is embraced by the cities worldwide and has influenced the existing transit service, especially the metro system. However, the urban regional mobility of FFB under the mixed relationship with metro is still unclear. In this study, we address this issue based on the three-month weekday FFB trips data in Shenzhen. First, we quantify to what extent the FFB competes with, connects with and complements metro system, and identify the corresponding travel scenarios. Then, the scenario-oriented travel communities that can reflect the distinctive urban regional mobility are detected. From the result, four kinds of travel scenarios are recognized, where the difference of metro service coverage towards access and egress FFB trips is further discussed. Moreover, by analyzing the development of the travel communities, we reveal the spatial and temporal characteristics of urban regional mobility. The findings enhance the understanding about the travel behaviors of FFB-metro integrated system, and can improve the daily operation of the transportation system. • Travel scenarios of free-floating bike (FFB)-metro integrated system are analyzed. • Egress coverage from metro tends to be larger than that of access to metro for FFB. • The dynamic development of regional mobility among urban regions is revealed. • The majority of FFB trips are found to happen in travel community regions. • The structure of travel communities helps to improve the operation of FFB. [ABSTRACT FROM AUTHOR]

• A multi-community spatio-temporal graph convolutional network is proposed to predict passenger demand. • The spatial correlations among regions are encoded through the graph convolutional network. • The Louvain algorithm is used to accomplish the passenger demand prediction of multi-regions. • The taxi order data collected in two cities are used in model validation and comparison. • The experiment result illustrates the superiority of the proposed model. Region-level passenger demand prediction plays an important role in the coordination of travel demand and supply in the urban public transportation system. The complex urban road network structure leads to irregular shapes and arrangements of regions, which poses a challenge for capturing the spatio-temporal correlation of demand generated in different regions. In this study, we propose a multi-community spatio-temporal graph convolutional network (MC_STGCN) framework to predict passenger demand at a multi-region level by exploring spatio-temporal correlations among regions. Specifically, the gated recurrent unit (GRU) is applied to encode the temporal correlation in regions into a vector. On the other hand, the spatial correlations among regions are encoded into two graphs through the graph convolutional network (GCN): geographically adjacent graph and functional similarity graph. Then, a prediction module based on the Louvain algorithm is used to accomplish the passenger demand prediction of multi-regions. The two real-world taxi order data collected in Shenzhen City and New York City are used in model validation and comparison. The numerical results show that the MC_STGCN model outperforms both classical time-series prediction methods and deep learning approaches. Moreover, in order to better illustrate the superiority of the proposed model, we further discuss the improvement of prediction performance though spatio-temporal correlation modeling and analyzing, the effectiveness of community detection compared with random classification of regions, and the advantages of regional level prediction compared with grid-based prediction models. [ABSTRACT FROM AUTHOR]