Search

• A non-linear programming model to design full-day demand-adapted timetables is proposed. • The time-dependent maximum sectional passenger flow is used as input data. • A new algorithm is proposed to match demand and services' supply. • The algorithm uses a rolling-horizon mechanism to iteratively compute departure times. • The effectiveness of the proposed algorithm is illustrated using real data of Chengdu Metro line 1. This paper focuses on the problem of designing the schedule of urban rail transit systems under dynamic passenger demand considering limited train capacity and congestion. From the point of view of operators, the goal of timetable optimization consists of using as few as possible service trains to transport the arriving passengers to their destinations securely and quickly. On the other hand, passengers want to spend the minimum time travelling to their destinations, including the time they wait on platforms until they board the first and successive trains in case of transferring between lines. Due to the variability of passenger flows and the difficulty in exactly solving full-day timetabling problems in short computation times, it is difficult in practice to use the approaches proposed in the literature. Moreover, when the capacity of trains is considered and oversaturation conditions emerge, the problem becomes even more complex, converting into a challenge the design of schedules from an operational point of view. By considering fixed train dwell times at stations and train running times between stations, we first propose a non-linear programming formulation to model the problem of determining the most convenient interdeparture times of services at the first station of a line, aiming at minimizing the waiting time of passengers. Since the dynamic behaviour of passenger demand and the maximum sectional passenger flow are approximately synchronous, an innovative method, the spatiotemporal synchronization coupling algorithm, based on working with the time-varying maximum sectional passenger flow instead of the demand patterns is proposed to quickly solve in an approximate way the demand-driven timetabling problem. The proposed approach is applied to a real case of Chengdu Metro Line 1 in China. The results show the effectiveness of the algorithm, both in terms of computational effort and in terms of reducing the waiting time currently experienced by passengers. [ABSTRACT FROM AUTHOR]

• A hybrid deep learning model within decomposition-ensemble strategy named DEASeq2Seq is proposed for multistep metro passenger flow prediction. • DEASeq2Seq employs CEEMDAN and RQA to capture intrinsic dynamics. • An attention-based Seq2Seq model is developed to characterize temporal dependencies. • Real-world datasets to verify the proposed model and further model interpretation. Short-term passenger flow prediction has practical significance for metro management and operation. However, the complex nonlinear and non-stationary characteristics make it challenging to detect evolution characteristics of passenger flow. To address this problem, a hybrid short-term metro passenger flow prediction model named decomposition ensemble attention sequence to sequence (DEASeq2Seq) is proposed in this paper. The proposed DEASeq2Seq includes three phases: decomposition, ensemble, and prediction. First, complete empirical mode decomposition with adaptive noise (CEEMDAN) is utilized to decompose the original passenger flow data into several intrinsic mode functions (IMFs) and a residue. Second, recurrence quantification analysis (RQA) is performed to reconstruct the decomposed modes into a stochastic part, a deterministic part, and a trend part via determinism evaluation. Third, a Seq2Seq model with the attention mechanism is proposed to execute multistep prediction for short-term passenger flow and explore the influence mechanism of the reconstructed components on the prediction targets. The real dataset from Chengdu metro, China, is used to verify the proposed model. The experiment results show that the proposed DEASeq2Seq model outperforms the benchmark models. Further model interpretations are conducted to analyze the impacts of decomposition strategy, ensemble strategy, and attention mechanism. [ABSTRACT FROM AUTHOR]