With the rapid development of China's maritime industry, accurate prediction of ship traffic flow is of great significance to improve the operational efficiency of ports. In this paper, we propose a combined model that combines the Seasonal Autoregressive Integrated Moving Average (SARIMA) model and Long Short-Term Memory (LSTM) fusing Multi-head Attention mechanism to improve the prediction accuracy of ship traffic flow. The example data summarizes the actual data from Guangzhou port production business system from 2014-2023. The study integrates the predictive capabilities of linear and nonlinear models while considering the influence of external economic factors on ship traffic flow, possessing better forecasting outcomes. Through comparison experiments with other three models, the combined deep learning model shows superior forecasting performance.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A Combined Deep Learning Method for Prediction of Ship Traffic Flow


    Contributors:
    Chen, Mengjiao (author) / Huang, Lei (author) / Wang, Ying (author)


    Publication date :

    2025-05-23


    Size :

    2814992 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Traffic flow prediction method based on deep learning

    XIE GANG / WANG HAIYING / XIE RUIQI | European Patent Office | 2024

    Free access

    Regional ship traffic flow prediction method and system

    CHEN XINQIANG / HAN BING / WU ZHONGDAI et al. | European Patent Office | 2023

    Free access

    Short-Term Traffic Flow Prediction: A Method of Combined Deep Learnings

    Chuanxiang Ren / Chunxu Chai / Changchang Yin et al. | DOAJ | 2021

    Free access

    Urban traffic flow prediction method based on deep learning

    WANG RONGXIU / DING YANQIU / ZHENG YAQIAN | European Patent Office | 2025

    Free access

    Lightweight traffic flow prediction method based on deep learning

    ZHENG RUI / DONG CHUNJIAO / GENG QINGQIAO et al. | European Patent Office | 2024

    Free access