Abstract
Effective management of urban rail transit systems increasingly centers on short-term passenger flow prediction which is a key factor in operational scheduling decisions. The critical need to use existing infrastructure efficiently and avoid potential emergencies due to large crowd gatherings. The advancements in short-term passenger flow forecasting have become essential within the intelligent transportation system domain. Despite its importance, there is a prominent gap in research specifically addressing the prediction of various types of passenger flows in subway systems. To bridge the gap, the paper proposed an innovative integrated method combining CNN-LSTM and SVM with Wavelet Transform harnessing their collective strengths. The approach involves a multi-stage process. Initially, passenger flow data is dissected into high and low-frequency series using wavelet transform. In the prediction phase, CNN-LSTM and SVM techniques are employed to, respectively learn and forecast these high and low-frequency sequence. The final phase involves the reintegration of these various predicted sequences through wavelet transform. The integrated approach aims to provide a more accurate prediction of short-term passenger flows in urban rail transit catering to the specific demands of the complex and dynamic field. The analysis shows that the SVM based Wavelet Transform method outperforms the CNN-LSTM based Wavelet Transform method in terms of efficiency and accuracy. Both approaches achieve better results than classical prediction techniques for passenger flow prediction on evaluation metrics, i.e., MAPE (9.80% and 7.28%), VAPE (0.78% and 3.85%), RMSE (2.48% and 0.76%), MSE (14.86% and 4.65%), R2 (1.57% and 12.57%) and accuracy (85.00% and 90.00%) respectively.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
Inquiries about data availability should be addressed to the authors.
References
Abbas R, Gu N (2023) Improving deep learning-based image super-resolution with residual learning and perceptual loss using SRGAN model. Soft Comput pp 1–17
Chen Q, Song Y, Zhao J (2021) Short-term traffic flow prediction based on improved wavelet neural network. Neural Comput Appl 33:8181–8190
Chuwang DD, Chen W, Zhong M (2023) Short-term urban rail transit passenger flow forecasting based on fusion model methods using univariate time series. Appl Soft Comput 147:110740
Filom S, Amiri AM, Razavi S (2022) Applications of machine learning methods in port operations–a systematic literature review. Transp Res Part E: Log Transp Rev 161:102722
Guo J, Wang W, Tang Y, Zhang Y, Zhuge H (2022) A CNN-Bi_LSTM parallel network approach for train travel time prediction. Knowl-Based Syst 256:109796
Hu G, Whalin RW, Kwembe TA, Lu W (2023) Short-term traffic flow prediction based on secondary hybrid decomposition and deep echo state networks. Physica A: Stat Mech Appl p 129313
Kanavos A, Kounelis F, Iliadis L, Makris C (2021) Deep learning models for forecasting aviation demand time series. Neural Comput Appl 33(23):16329–16343
Li X, Zhang Y, Du M, Yang J (2020) The forecasting of passenger demand under hybrid ridesharing service modes: a combined model based on WT-FCBF-LSTM. Sustain Cities Soc 62:102419
Li H, Jin K, Sun S, Jia X, Li Y (2022a) Metro passenger flow forecasting though multi-source time-series fusion: an ensemble deep learning approach. Appl Soft Comput 120:108644
Li W, Guan H, Han Y, Zhu H, Wang A (2022b) Short-term holiday travel demand prediction for urban tour transportation: a combined model based on STC-LSTM deep learning approach. KSCE J Civ Eng 26(9):4086–4102
Li P, Wu W, Pei X (2023) A separate modelling approach for short-term bus passenger flow prediction based on behavioural patterns: a hybrid decision tree method. Physica A 616:128567
Liu M, Wang W, Hu X, Fu Y, Xu F, Miao X (2023) Multivariate long-time series traffic passenger flow prediction using causal convolutional sparse self-attention MTS-Informer. Neural Computing and Applications, pp.1–17.
Luo B, Cao R, Yang W, Wang Z, Hu X, Xu J, Fan Z, Zhang L (2023) Analysing and predicting the fine-scale distribution of traffic particulate matter in urban nonmotorized lanes by using wavelet transform and random forest methods. Stochastic Environmental Research and Risk Assessment, pp 1–20
Lv S, Wang K, Yang H, Wang P (2024) An origin–destination passenger flow prediction system based on convolutional neural network and passenger source-based attention mechanism. Expert Syst Appl 238:121989
Mulerikkal, J., Thandassery, S., Rejathalal, V. and Kunnamkody, D.M.D., 2022. Performance improvement for metro passenger flow forecast using spatio-temporal deep neural network. Neural Computing and Applications, pp.1–12.
Nagaraj N, Gururaj HL, Swathi BH, Hu YC (2022) Passenger flow prediction in bus transportation system using deep learning. Multimed Tools Appl 81(9):12519–12542
Narmadha S, Vijayakumar V (2021) Spatio-Temporal vehicle traffic flow prediction using multivariate CNN and LSTM model. Materials today: proceedings.
Shen, J., 2022, January. Rail Passenger Flow Prediction Combining Social Media Data for Rail Passenger. In LISS 2021: Proceedings of the 11th International Conference on Logistics, Informatics and Service Sciences (pp. 672–681). Singapore: Springer Nature Singapore.
Wang X, Xu X, Wu Y, Liu J (2022) An effective spatiotemporal deep learning framework model for short-term passenger flow prediction. Soft Comput 26(12):5523–5538
Xiu C, Sun Y, Peng Q, Chen C, Yu X (2022) Learn traffic as a signal: Using ensemble empirical mode decomposition to enhance short-term passenger flow prediction in metro systems. J Rail Transp Plan Manag 22:100311
Xue Q, Zhang W, Ding M, Yang X, Wu J, Gao Z (2023) Passenger flow forecasting approaches for urban rail transit: a survey. Int J Gen Syst 52(8):919–947
Yang X, Xue Q, Yang X, Yin H, Qu Y, Li X, Wu J (2021) A novel prediction model for the inbound passenger flow of urban rail transit. Inf Sci 566:347–363
Yang H, Cheng Y, Li G (2022) A new traffic flow prediction model based on cosine similarity variational mode decomposition, extreme learning machine and iterative error compensation strategy. Eng Appl Artif Intell 115:105234
Yu J (2022) Short-term airline passenger flow prediction based on the attention mechanism and gated recurrent unit model. Cogn Comput 14(2):693–701
Zhou K, Oh SK, Pedrycz W, Qiu J (2023) Data preprocessing strategy in constructing convolutional neural network classifier based on constrained particle swarm optimization with fuzzy penalty function. Eng Appl Artif Intell 117:105580
Funding
This study was funded by Soft Science Project of Science and Technology Department of Henan Province: Research on the Construction and Competitiveness Promotion of Cultural Tourism Circle in the Middle Reaches of the Yellow River, Project No: 222400410363.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
No conflict of interest has been declared by the authors.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Xu, Q. Incorporating CNN-LSTM and SVM with wavelet transform methods for tourist passenger flow prediction. Soft Comput 28, 2719–2736 (2024). https://doi.org/10.1007/s00500-023-09592-w
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-023-09592-w