Predictability of short-term passengers’ origin and destination demands in urban rail transit

Yang, Fang; Shuai, Chunyan; Qian, Qian; Wang, Wencong; He, Mengchang; Lee, Jae Young

doi:10.1007/s11116-022-10313-9

Cited by 6 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an emerging technology, deep learning has been widely used in various fields with its powerful feature extraction and expression ability. In recent years, several deep learning methods have emerged for short-term OD prediction in URT [8,[17][18][19][20][21][22]. Based on the multisource spatio-temporal data, Ref.…”

Section: Deep Learning Methods For Short-term Od Prediction In Urban ...mentioning

confidence: 99%

See 1 more Smart Citation

Parallel Interactive Attention Network for Short-Term Origin–Destination Prediction in Urban Rail Transit

Zhou,

Gao,

Tang

2023

Applied Sciences

View full text Add to dashboard Cite

Short-term origin–destination (termed as OD) prediction is crucial to improve the operation of urban rail transit (termed as URT). The latest research results show that deep learning can effectively improve the performance of short-term OD prediction and meet the real-time requirements. However, many advanced neural network design ideas have not been fully applied in the field of short-term OD prediction in URT. In this paper, a novel parallel interactive attention network (termed as PIANet) for short-term OD prediction in URT is proposed to further improve the short-term OD prediction accuracy. In the proposed PIANet, a novel omnidirectional attention module (termed as OAM) is proposed to improve the representational power of the network by calculating the feature weights in the channel–spatial dimension. Moreover, a simple yet effective feature interaction is proposed to improve the feature utilization. Based on the two real-world datasets from the Beijing subway, the comparative experiments demonstrate that the proposed PIANet outperforms the state-of-the-art deep learning methods for short-term OD prediction in URT, and the ablation studies demonstrate that the proposed OAMs and feature interaction play an important role in improving the short-term OD prediction accuracy.

show abstract

Section: Deep Learning Methods For Short-term Od Prediction In Urban ...mentioning

confidence: 99%

“…Ref. [20] proposed temporal Pearson correlation coefficients, approximate entropy, and spatial correlations as indicators to reflect the inherent spatio-temporal correlations and complexity of the OD flow. Ref.…”

Section: Deep Learning Methods For Short-term Od Prediction In Urban ...mentioning

confidence: 99%

Parallel Interactive Attention Network for Short-Term Origin–Destination Prediction in Urban Rail Transit

Zhou,

Gao,

Tang

2023

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Furthermore, the adopted interval is relatively stable in order not to conceal the timevarying laws of passenger demand. The review showed that 15 min and 60 min were the preferred time intervals for data aggregation (Giraldo-Forero et al, 2019; Yang et al, 2022).…”

Section: What Are the Methods For Collecting Passenger Data?mentioning

confidence: 99%

Estimating Passenger Demand Using Machine Learning Models: A Systematic Review

Boateng,

Anum Adams,

Kofi Akowuah

2023

E3S Web Conf.

View full text Add to dashboard Cite

This article investigated machine learning models used to estimate passenger demand. These models have the potential to provide valuable insights into passenger trip behaviour and other inferences. The estimate of passenger demand using machine learning model research and the methodologies used are fragmented. To synchronise these studies, this paper conducts a systematic review of machine learning models to estimate passenger demand. The review investigates how passenger demand is estimated using machine learning models. A comprehensive search strategy is conducted across the three main online publishing databases to locate 911 unique records. Relevant record titles, abstracts, and publication information are extracted, leaving 102 articles. Furthermore, articles are evaluated according to eligibility requirements. This procedure yields 21 full-text papers for data extraction. 3 research thematic questions covering passenger data collection techniques, passenger demand interventions, and intervention performance are reviewed in detail. The results of this study suggest that mobility records, LSTM-based models, and performance metrics play a critical role in conducting passenger demand prediction studies. The model evaluation was mostly restricted to 3 performance metrics which needs improved metric for evaluation. Furthermore, the review determined an overreliance on the longand short-term memory model to estimate passenger demand. Therefore, minimising the limitation of the LSTM model will generally improve the estimation models. Furthermore, having an acceptable trainset to avoid overfitting is crucial. In addition, it is advisable to consider multiple metrics to have a more comprehensive evaluation.

show abstract

“…Several scholars have employed various models such as the least squares method [11], structural state-space models [12][13][14][15], maximum entropy models [16], and dynamic mode decomposition [17] to estimate and predict time-varying OD matrices. Another category of methods relies on probability model and statistical analysis, the concept of maximum probable relative error (MPRE) [18] and encompassing Bayesian analysis [19][20][21]. Deep learning techniques have emerged as a promising field in OD prediction.…”

Section: Introductionmentioning

confidence: 99%

Origin-Destination prediction from Road Segment Speed Data using GraphResLSTM Model

Zhang,

2024

Preprint

View full text Add to dashboard Cite

Amidst the escalating demands for efficient Intelligent Transportation Systems (ITS) management, the significance of accurate Origin-Destination (OD) data prediction has emerged as a paramount concern, given its indispensable role in the ITS domain. This paper introduces a hybrid model, GraphResLSTM, which integrates Graph Convolutional Network, Residual Neural Network, and Long Short-Term Memory Network, to exploit road segment average speed data for OD prediction. Compared with conventional road segment traffic volume data, average speed data is more readily available, significantly reducing the data accessibility barrier for OD prediction. We underscore the critical role of selection of key road segments and employs a combination of Entropy Weight Method and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to identify 41 representative road segments from the experimental road network. In the experimental phase, we simulate 86400000 seconds of traffic conditions, generating a total of 1440000 data records. Utilizing these simulated data, we design and conduct comparative experiments between models as well as data types. The results demonstrate that both the GraphResLSTM model and road segment average speed data outperform other models and data types notably in OD prediction tasks.

show abstract

Predictability of short-term passengers’ origin and destination demands in urban rail transit

Cited by 6 publications

References 38 publications

Parallel Interactive Attention Network for Short-Term Origin–Destination Prediction in Urban Rail Transit

Parallel Interactive Attention Network for Short-Term Origin–Destination Prediction in Urban Rail Transit

Estimating Passenger Demand Using Machine Learning Models: A Systematic Review

Origin-Destination prediction from Road Segment Speed Data using GraphResLSTM Model

Contact Info

Product

Resources

About