Journal of Jilin University(Engineering and Technology Edition) ›› 2023, Vol. 53 ›› Issue (7): 2001-2015.doi: 10.13229/j.cnki.jdxbgxb.20210935

Previous Articles    

Deep learning method for bus passenger flow prediction based on multi-source data and surrogate-based optimization

Wei-tiao WU1(),Kun ZENG1,Wei ZHOU1,Peng LI2(),Wen-zhou JIN1   

  1. 1.School of Civil Engineering and Transportation,South China University of Technology,Guangzhou 510641,China
    2.School of Automotive and Transportation Engineering,Shenzhen Polytechnic,Shenzhen 518055,China
  • Received:2021-09-16 Online:2023-07-01 Published:2023-07-20
  • Contact: Peng LI E-mail:ctwtwu@scut.edu.cn;lipeng@szpt.edu.cn

Abstract:

A scalable deep learning framework is proposed to leverage multi-source data in bus passenger flow prediction. First, on the basis of four external factors, three internal factors are introduced as explanatory variables of passenger flow of a bus route. The variance reduction method is used to verify the linkage relationship between internal factors and external factors, and the necessity of capturing the linkage relationship of multi-source data. Then, by utilizing the advantages of convolutional operations of convolutional neural networks in handling two-dimensional data, the influential factors of passenger flow are visualized and an hourly passenger flow subdivision matrix is constructed to adapt to convolutional operations, such that the linkage between multi-source data is captured. To further improve the prediction performance, the matrix structure optimization problem is transformed into a Traveling Salesman Problem, and the surrogate-based optimization technique is used to efficiently optimize the structure of hourly passenger flow subdivision matrix. Finally, the data from Route-281 bus in Guangzhou, China was used as an example for validation. The results show that, through optimizing the structure of hourly passenger flow subdivision matrix, the prediction accuracy can be effectively improved, and optimal usage of data resources can be achieved. The independent effect of internal factors was not significant, whereas the combined effect of external and internal factors is significant. In terms of prediction accuracy,the proposed method has certain advantages compared to the results only considering external factors and other deep learning models.

Key words: engineering of communication and transportation system, passenger flow prediction, deep learning, bus passenger flow, multi-source data, surrogate-based optimization

CLC Number: 

  • U491.1

Fig.1

Flow chart of research process"

Fig.1

Passenger flow distribution on a typical day"

Table 1

Analysis of passenger trip pattern"

类别编号刷卡时间刷卡类型行为特征
115∶00~19∶001下午下班出行
210∶00~14∶003老人中午出行
315∶00~18∶002学生下午放学出行
46∶00~10∶001上午上班出行
56∶00~10∶002学生上午上学出行
619∶00~23∶002学生晚上其他出行
720∶00~23∶003老人晚上其他出行
820∶00~23∶001晚上其他出行
96∶00~9∶003老人上午出行
1011∶00~14∶001中午上下班出行
1115∶00~19∶003老人下午出行
1211∶00~14∶002学生中午上下学出行

Table 2

Classification of dependence on line of passengers"

周乘车次数标准差周平均乘车次数类别标准:周乘坐时间的标准差和平均值(min)乘客对公交线路依赖性
第1类标准差<1.418且平均值<3.136出行次数少连续性好
第2类标准差<1.418且平均值≥3.136出行次数多连续性好
第3类标准差≥1.418且平均值<3.136出行次数少连续性差
第4类标准差≥1.418且平均值≥3.136出行次数多连续性差

Fig.3

Clustering results for four months with respectto line stickiness"

Fig.4

Relative contribution of internal factors to external factors"

Fig.5

Construction of hourly passenger flow subdivision matrix"

Table 3

Definitions of columns in hourly passenger flow subdivision matrix"

客流特征向量定义客流特征向量定义
x1刷卡时间x5天气因素
x2线路依赖x6是否为高峰时段
x3刷卡类型x7行为模式
x4工作日性质x8历史客流

Table 4

Test samples of passenger subdivision matrices"

结构编号结构
Originalx1,x2,x3,x4,x5,x6,x7,x8
1x5,x4,x1,x6,x7,x3,x8,x2
2x3,x1,x2,x4,x6,x5,x7,x8
3x3,x8,x1,x6,x4,x2,x5,x7

Fig.6

Heat maps of correlation coefficients of fourhourly passenger flow subdivision matrice"

Fig.7

Illustration of matrix structures and the corresponding MAPEs"

Fig.8

Flow chart of surrogate-based optimization"

Fig.9

281 Bus route in Guangzhou"

Fig.10

Prediction results of four schemes"

Fig.11

Prediction performance of CNN and Conv-LSTM for different schemes"

Fig.12

Relative contribution of influential factorsbased on RMSE increase"

Table 5

Comparisons of rolling prediction performance"

单步向前二步向前三步向前
MAPE /%RMSEMAPE/%RMSEMAPE/%RMSE
Conv?LSTM9.70223.4210.13230.1510.40232.67
CNN9.65226.0210.06234.4610.27236.72
LSTM11.82259.7111.75258.4311.94262.73
MLP16.31470.8322.31737.7725.27823.54
注意力MLP10.86246.8511.34258.5011.72266.18

Fig.13

Prediction of passenger flow of 3 card types"

Fig.14

Prediction of passenger flow of different levels of line stickiness"

Fig.15

Prediction of passenger flow of 12 types of travel patterns"

1 Wei Y, Chen M C. Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks[J]. Transportation Research Part C, 2012, 21:148-162.
2 Liu Y, Liu Z Y, Jia R. DeepPF: a deep learning based architecture for metro passenger flow prediction[J]. Transportation Research Part C, 2019, 101: 18-34.
3 张春辉, 宋瑞, 孙杨. 基于卡尔曼滤波的公交站点短时客流预测[J]. 交通运输系统工程与信息, 2011, 11(4):154-159.
Zhang Chun-hui, Song Rui, Sun Yang. Kalman filter-based short-term passenger flow forecasting on bus stop[J]. Journal of Transportation Systems Engineering and Information Technology, 2011, 11(4): 154-159.
4 王庆荣, 张秋余. 基于随机灰色蚁群神经网络的近期公交客流预测[J]. 计算机应用研究, 2012, 29(6): 2078-2080.
Wang Qing-rong, Zhang Qiu-yu. Forecasting of short-term urban public transit volume based on random gray ant colony neural network[J]. Application Research of Computers, 2012, 29(6):2078-2080.
5 王立政, 朱从坤. 基于LM-BP算法的轨道交通客流短时预测[J]. 价值工程, 2018(3):154-156.
Wang Li-zheng, Zhu Cong-kun. Short-term prediction research on urban rail transit passenger flow based on LM-BP algorithm[J]. Value Engineering, 2018(3):154-156.
6 Ma X L, Wu Y J, Wang Y H, et al. Mining smart card data for transit riders' travel patterns[J]. Transportation Research Part C, 2013, 36:1-12.
7 Chen X Q, Xiong C F, He X, et al. Time-of-day vehicle mileage fees for congestion mitigation and revenue generation: a simulation-based optimization method and its real-world application[J]. Transportation Research Part C, 2016, 63: 71-95.
8 Wu W T, Li P, Liu R H, et al. Predicting peak load of bus routes with supply optimization and scaled Shepard interpolation: a newsvendor model[J]. Transportation Research Part E, 2020, 142:No.102041.
9 Yu B, Guo Z, Asian S, et al. Flight delay prediction for commercial air transport: a deep learning approach[J]. Transportation Research Part E, 2019, 125: 203-221.
[1] Zhen-hai ZHANG,Kun JI,Jian-wu DANG. Crack identification method for bridge based on BCEM model [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(5): 1418-1426.
[2] Chao SUN,Hao-wei YIN,Wen-yun TANG,Zhao-ming CHU. Sensor deployment strategy and expansion inference of mobile phone data for traffic demand estimation [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(4): 1070-1077.
[3] Fei WU,Hao-ye NONG,Chen-hao MA. Tool wear prediction method based on particle swarm optimizationlong and short time memory model [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(4): 989-997.
[4] Wen-li JI,Zhong TIAN,Jing CHAI,Ding-ding ZHANG,Bin WANG. Prediction of water⁃flowing height in fractured zone based on distributed optical fiber and multi⁃attribute fusion [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(4): 1200-1210.
[5] Hong-fei JIA,Ying-jun XU,Li-li YANG,Nan WANG. League member selection and benefit distribution of commercial vehicles multi⁃modal transportation [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(4): 1060-1069.
[6] Ke HE,Hai-tao DING,Xuan-qi LAI,Nan XU,Kong-hui GUO. Wheel odometry error prediction model based on transformer [J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(3): 653-662.
[7] Jin-wu GAO,Zhi-huan JIA,Xiang-yang WANG,Hao XING. Degradation trend prediction of proton exchange membrane fuel cell based on PSO⁃LSTM [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(9): 2192-2202.
[8] Xian-min SONG,Shu-tian YANG,Ming-xin LIU,Zhi-hui LI. Fluctuation characteristics and prediction method of bus travel time between stations [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1792-1799.
[9] Xiao-ying LI,Ming YANG,Rui QUAN,Bao-hua TAN. Unbalanced text classification method based on deep learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1889-1895.
[10] Xuan-jing SHEN,Xue-feng ZHANG,Yu WANG,Yu-bo JIN. Multi⁃focus image fusion algorithm based on pixel⁃level convolutional neural network [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1857-1864.
[11] Dan HU,Xin MENG. Vessel search method by earth observation satellite based on time⁃varying grid [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(8): 1896-1903.
[12] Wen-jing WU,Yong-bin ZHAN,Li-li YANG,Run-chao CHEN. Coordinated control method of variable speed limit in on⁃ramp area considering safety distance [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1315-1323.
[13] Zhi ZHENG,Bo GENG,Fu-min WANG,Jun-hong DONG,Si-si WEI. Improvement of protective ability for existing low⁃grade concrete guardrail [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1362-1374.
[14] Ming-hua GAO,Can YANG. Traffic target detection method based on improved convolution neural network [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(6): 1353-1361.
[15] Shi-min FANG. Multiple source data selective integration algorithm based on frequent pattern tree [J]. Journal of Jilin University(Engineering and Technology Edition), 2022, 52(4): 885-890.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!