Open AccessResearch on Container Throughput Forecast Based on ARIMA-BP Neural Network
Author(s) -
Yifei Zhang,
Yang Fu,
Genghua Li
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1634/1/012024
Subject(s) - autoregressive integrated moving average , artificial neural network , mean squared error , mean absolute percentage error , throughput , residual , container (type theory) , computer science , weighting , statistics , data mining , artificial intelligence , algorithm , machine learning , mathematics , engineering , time series , telecommunications , mechanical engineering , medicine , wireless , radiology
In order to improve the accuracy of the container throughput, the paper proposed a prediction method based on ARIMA-BP neural network for the container throughput, and compared with the combined prediction method based on ARIMA-BP neural network, from the perspective of simple weighting and residual optimization. It is applied to the container throughput prediction of the Qingdao port statistics for a total of 24 quarters from 2014-2019. The results show that the prediction accuracy of the combination prediction method based on residual optimization was the highest. Compared with other prediction models, the evaluation indexes RMSE(Root Mean Square Error), MAE(Mean Absolute Error), and MAPE(Mean Absolute Percentage Error) were 15.95, 13.31 and 2.52% respectively and the prediction accuracy based on the BP neural network was lowest. The prediction method proposed in this paper for container throughput can provide guidance for the related personnel.