A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit
Abstract Short-term passenger flow prediction (STPFP) helps ease traffic congestion and optimize the allocation of rail transit resources. However, the nonlinear and nonstationary nature of passenger flow time series challenges STPFP. To address this issue, a hybrid model based on time series decomp...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-11-01
|
| Series: | Urban Rail Transit |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40864-023-00205-1 |
| _version_ | 1797398139735375872 |
|---|---|
| author | Jinxin Wu Deqiang He Xianwang Li Suiqiu He Qin Li Chonghui Ren |
| author_facet | Jinxin Wu Deqiang He Xianwang Li Suiqiu He Qin Li Chonghui Ren |
| author_sort | Jinxin Wu |
| collection | DOAJ |
| description | Abstract Short-term passenger flow prediction (STPFP) helps ease traffic congestion and optimize the allocation of rail transit resources. However, the nonlinear and nonstationary nature of passenger flow time series challenges STPFP. To address this issue, a hybrid model based on time series decomposition and reinforcement learning ensemble strategies is proposed. Firstly, the improved arithmetic optimization algorithm is constructed by adding sine chaotic mapping, a new dynamic boundary strategy, and adaptive T distribution mutations for optimizing variational mode decomposition (VMD) parameters. Then, the original passenger flow data containing nonlinear and nonstationary irregular changes of noise is decomposed into several intrinsic mode functions (IMFs) by using the optimized VMD technology, which reduces the time-varying complexity of passenger flow time series and improves predictability. Meanwhile, the IMFs are divided into different frequency series by fluctuation-based dispersion entropy, and diverse models are utilized to predict different frequency series. Finally, to avoid the cumulative error caused by the direct superposition of each IMF’s prediction result, reinforcement learning is adopted to ensemble the multiple models to acquire the multistep passenger flow prediction result. Experiments on four subway station passenger flow datasets proved that the prediction performance of the proposed method was better than all benchmark models. The excellent prediction effect of the proposed model has important guiding significance for evaluating the operation status of urban rail transit systems and improving the level of passenger service. |
| first_indexed | 2024-03-09T01:20:25Z |
| format | Article |
| id | doaj.art-5bad901977d74f1a87b08d6d3b938fc2 |
| institution | Directory Open Access Journal |
| issn | 2199-6687 2199-6679 |
| language | English |
| last_indexed | 2024-03-09T01:20:25Z |
| publishDate | 2023-11-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Urban Rail Transit |
| spelling | doaj.art-5bad901977d74f1a87b08d6d3b938fc22023-12-10T12:10:44ZengSpringerOpenUrban Rail Transit2199-66872199-66792023-11-019432335110.1007/s40864-023-00205-1A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail TransitJinxin Wu0Deqiang He1Xianwang Li2Suiqiu He3Qin Li4Chonghui Ren5Guangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology, School of Mechanical Engineering of Guangxi UniversityGuangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology, School of Mechanical Engineering of Guangxi UniversityGuangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology, School of Mechanical Engineering of Guangxi UniversityGuangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology, School of Mechanical Engineering of Guangxi UniversityGuangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology, School of Mechanical Engineering of Guangxi UniversityNanning Rail Transit Co., Ltd.Abstract Short-term passenger flow prediction (STPFP) helps ease traffic congestion and optimize the allocation of rail transit resources. However, the nonlinear and nonstationary nature of passenger flow time series challenges STPFP. To address this issue, a hybrid model based on time series decomposition and reinforcement learning ensemble strategies is proposed. Firstly, the improved arithmetic optimization algorithm is constructed by adding sine chaotic mapping, a new dynamic boundary strategy, and adaptive T distribution mutations for optimizing variational mode decomposition (VMD) parameters. Then, the original passenger flow data containing nonlinear and nonstationary irregular changes of noise is decomposed into several intrinsic mode functions (IMFs) by using the optimized VMD technology, which reduces the time-varying complexity of passenger flow time series and improves predictability. Meanwhile, the IMFs are divided into different frequency series by fluctuation-based dispersion entropy, and diverse models are utilized to predict different frequency series. Finally, to avoid the cumulative error caused by the direct superposition of each IMF’s prediction result, reinforcement learning is adopted to ensemble the multiple models to acquire the multistep passenger flow prediction result. Experiments on four subway station passenger flow datasets proved that the prediction performance of the proposed method was better than all benchmark models. The excellent prediction effect of the proposed model has important guiding significance for evaluating the operation status of urban rail transit systems and improving the level of passenger service.https://doi.org/10.1007/s40864-023-00205-1Variational mode decompositionReinforcement learningEnsemble strategyUrban rail transitShort-term passenger flow prediction |
| spellingShingle | Jinxin Wu Deqiang He Xianwang Li Suiqiu He Qin Li Chonghui Ren A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit Urban Rail Transit Variational mode decomposition Reinforcement learning Ensemble strategy Urban rail transit Short-term passenger flow prediction |
| title | A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit |
| title_full | A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit |
| title_fullStr | A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit |
| title_full_unstemmed | A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit |
| title_short | A Time Series Decomposition and Reinforcement Learning Ensemble Method for Short-Term Passenger Flow Prediction in Urban Rail Transit |
| title_sort | time series decomposition and reinforcement learning ensemble method for short term passenger flow prediction in urban rail transit |
| topic | Variational mode decomposition Reinforcement learning Ensemble strategy Urban rail transit Short-term passenger flow prediction |
| url | https://doi.org/10.1007/s40864-023-00205-1 |
| work_keys_str_mv | AT jinxinwu atimeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT deqianghe atimeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT xianwangli atimeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT suiqiuhe atimeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT qinli atimeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT chonghuiren atimeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT jinxinwu timeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT deqianghe timeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT xianwangli timeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT suiqiuhe timeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT qinli timeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit AT chonghuiren timeseriesdecompositionandreinforcementlearningensemblemethodforshorttermpassengerflowpredictioninurbanrailtransit |