Reliable and accurate estimates of metro demand can provide metro authorities with insightful information for the planning of route alignment and station locations. Many existing studies focus on metro demand from daily or annual ridership profiles, but only a few concern the variation in hourly ridership. In this paper, a geographically and temporally weighted regression (GTWR) model was used to examine the spatial and temporal variation in the relationship between hourly ridership and factors related to the built environment and topological structure. Taking Nanjing, China as a case study, an empirical study was conducted with automatic fare collection (AFC) data in three weeks. With an analysis of variance (ANOVA), it was found that the GTWR model produced the best fit for hourly ridership data compared with traditional regression models. Four built-environment factors, namely residence, commerce, scenery, and parking, and two topological-structure factors, namely degree centrality and closeness centrality, were proven to be significantly related to station-level ridership. The spatial distribution pattern and temporal nonstationarity of these six variables were further analyzed. The result of this study confirmed that the GTWR model can provide more realistic and useful information by capturing spatiotemporal heterogeneity.

Forecasting for short-term ridership is the foundation of metro operation and management. A prediction model is necessary to seize the weekly periodicity and nonlinearity characteristics of short-term ridership in real-time. First, this research captures the inherent periodicity of ridership via seasonal autoregressive integrated moving average model (SARIMA) and proposes a support vector machine overall online model (SVMOOL) which insets the weekly periodic characteristics and trains the updated data day by day. Then, this research captures the nonlinear characteristics of the ridership via successive ridership value inputs and proposes a support vector machine partial online model (SVMPOL) which insets the nonlinear characteristics and trains the updated data of the predicted day by time interval (such as 5-min). Afterwards, to avoid the drawbacks and to take advantages of the strengths of the two individual online models, this research takes the average predicted values of two models as the final predicted values, which are called support vector machine combined online model (SVMCOL). Finally, this research uses the 5-min ridership at Zhujianglu and Sanshanjie Stations of Nanjing Metro to compare the SVMCOL model with three well-known prediction models including SARIMA, back-propagation neural network (BPNN), and SVM models. The resultant performance comparisons suggest that SARIMA is superior for the stable weekday ridership to other models. Yet the SVMCOL model is the best performer for the unstable weekend ridership and holiday ridership. It shows that for metro operation manager that gear toward timely response to real-world unstable and abnormal situations, the SVMCOL may be a better tool than the three well-known models.

With reliable and accurate predictions of short-term passenger flow, metro agencies can assign proper trains and crews, optimize schedules, and operate efficiently. This paper presents a hybrid model, which is intended for forecasting short-term passenger flow. The forecasting model aggregates SVR-based sub-models developed with subsets extracted by combining bootstrap sampling with dimensions subsampling. The sub-models consider not only the data from the reconstructed chaos attractor but also the periodic historical data. Passenger flow data collected from four different metro stations in Nanjing are used for model implementation and performance evaluation. Results confirm that the proposed EICSVR model was able to significantly improve predictive performance and generalization. The excellent accuracy and stability obtained in the empirical study indicate that the proposed model has good development potential for forecasting short-term passenger flow.

Promoting a transition in individuals’ travel mode from car to an integrated metro and bikeshare systems is expected to effectively reduce the traffic congestion that results mainly from commute trips performed by individual automobiles. This paper focuses on the use frequency of an integrated metro–bikeshare by individuals, and presents empirical evidence from Nanjing, China. Using one-week GPS data collected from the Mobike company, the spatiotemporal characteristics of origin/destination for cyclists who would likely to use shared bike as a feeder mode to metro are examined. Three areas of travel-related spatiotemporal information were extracted including (1) the distribution of walking distances between metro stations and shared bike parking lots; (2) the distribution of cycling times between origins/destinations and metro stations; and (3) the times when metro–bikeshare users pick up/drop off shared bikes to transfer to/from a metro. Incorporating these three features into a questionnaire design, an intercept survey of possible factors on the use of the combined mode was conducted at seven functional metro stations. An ordered logistic regression model was used to examine the significant factors that influence groupings of metro passengers. Results showed that the high-, medium- and low-frequency groups of metro–bikeshare users accounted for 9.92%, 21.98% and 68.1%, respectively. Education, individual income, travel purpose, travel time on the metro, workplace location and bike lane infrastructure were found to have significant impacts on metro passengers’ use frequency of integrated metro–bikeshares. Relevant policies and interventions for metro passengers of Nanjing are proposed to encourage the integration of metro and bikeshare systems.