Short-term Passenger Flow Prediction for Hangzhou Metro based on Machine Learning Algorithms
DOI:
https://doi.org/10.54097/fzedhe98Keywords:
Metro, LSTM, random forest, forecast.Abstract
Subways have now reached an irreplaceable position in urban rail transportation. The aggregation of people choosing subway travel at the same time can easily lead to passenger congestion, so predicting passenger flow in advance is crucial. This study used LSTM (Long Short-Term Memory) as the data prediction model to forecast the card-swiping data of the Hangzhou subway for 10 consecutive days. The historical data was divided into intervals of 3 minutes, and the random forest was used to compare the correlation among features. The final prediction results showed that both the inbound and outbound flows exhibited a bimodal pattern on weekdays, while on holidays, the prediction results showed a bimodal pattern. Furthermore, by adjusting the parameters appropriately, the simulated trends closely matched the actual values, indicating the applicability of LSTM in simulating short-term passenger flow. Accurate prediction of passenger flow can enable more comprehensive management of the subway system and provide passengers with a higher-quality service experience and travel options.
Downloads
References
Cheng Qiyun, Sun Caixin, Zhang Xiaoxing, et al. Short-Term load forecasting model and method for a power system based on complementation of neural network and fuzzy logic. Transactions of China Electrotechnical Society, 2004, 19(10): 53-58.
Chang, H. Short-Term Passenger Flow Prediction for Subway Based on LSTM Neural Network. Xi Jing University, 2022.
Wang, J. Research on Short-Term Passenger Flow Prediction for Urban Rail Transit Based on Deep Learning Models. (Doctoral dissertation). Beijing Jiaotong University, 2020.
Lv, H., Feng, Q. A Review of Random Forest Algorithm Research. Journal of Hebei Academy of Sciences, 2019, 36(03): 37-41.
Wang, Q. Research on Smart Tourism Based on Big Data Analysis. (Doctoral dissertation). Changchun University of Technology, 2021.
Feng, B. Research on Short-Term Passenger Flow Prediction for Urban Rail Transit Based on CNN-LSTM Combination Model. East China Jiaotong University, 2020.
Zhang, Z. Research on Short-Term Traffic Flow Prediction Based on Chaos Analysis and MEA-LSTM Model. Chang'an University, 2021.
Peng K, et al. Research on Passenger Flow Prediction of Railway Stations Based on Improved LSTM Model. Railway Transport and Ma Kunlong. Short term distributed load forecasting method based on big data. Changsha: Hunan University, 2014.
Zhang, J. Research on Short-Term Passenger Flow Prediction Methods for Urban Rail Transit Network. Beijing Jiaotong University, 2021.
BeAmjady N. Short-term hourly load forecasting using time series modeling with peak load estimation capability. IEEE Transactions on Power Systems, 2001, 16(4): 798-805.
Liu W, et al. Research on Holiday Passenger Flow Prediction in Suzhou Rail Transit. Urban Rapid Rail Transit, 2021, 34(05): 66-73.
Downloads
Published
Issue
Section
License
Copyright (c) 2023 Highlights in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
