Search

High-precision short-term inbound passenger flow prediction is of great significance to the daily crowd management and line rescheduling in urban rail systems. Although current models have been applied to prediction, most methods need optimization to meet refined passenger flow management demand. In order to better predict the passenger flow, a novel Wave-LSTM model, based on long short-term memory network (LSTM) and wavelet, is introduced in this paper. In an empirical study with practical passenger flow data of Dongzhimen Station in the Beijing Subway system, the hybrid model exhibited more effective performance in terms of prediction accuracy than the existing algorithms, e.g., autoregressive integrated moving average (ARIMA), nonlinear regression (NAR), and traditional LSTM model. The study illustrates that our newly adopted model is a promising approach for predicting high-precision short-term inbound passenger flow. [ABSTRACT FROM AUTHOR]

• GPS and IC card data are used to simulate the time-dependent travel time and passenger demand. • An improved passenger selection model is proposed by introducing a preference coefficient. • A working hours constraint is first formulated in bus timetabling problem. • The model is transformed into an integer linear program, followed by a model contraction approach. • A two-stage solution method is proposed for large-scale bus timetabling problems. Bus lines are suffering from serious decline in passenger volume due to the rapid development of urban rail transit and shared transport, and big data intelligence may help them change the status quo. However, the tremendous amount of travel data collected in recent years have not got effectively utilization. In order to improve passenger volume for bus lines, this paper devotes to develop a data-driven bus timetable to substitute the existing experience-based bus timetable, which is now widely used by bus lines. Driven by the bus GPS data and IC card data, a timetable optimization model with time-dependent passenger demand and travel time among stops is proposed. The objective of maximizing passenger volume is based on a new preference-based passenger selection model. The working hours constraint is initially formulated, and the headway constraint and departure time constraints are also taken into account. For handling the step functions in both objective and constraints, we introduce a set of 0–1 variables to transform the proposed model into an integer linear programming. A model contraction approach is provided for solving the medium-scale problems and a two-stage solution method is proposed for the large-scale problems. The proposed model and methodology are tested on a real-world bus line in Beijing. The results show that it is able to produce a satisfactory timetable that outperforms the previously used experience-based one in terms of raising the average passenger volume by 8.2%. [ABSTRACT FROM AUTHOR]