Ship traffic flow forecast of Qingdao port based on LSTM

Zhe Ji; Le Wang; Xiaobo Zhang; Fengwu Wang

doi:10.1117/12.2623859

6 February 2022 Ship traffic flow forecast of Qingdao port based on LSTM

Zhe Ji, Le Wang, Xiaobo Zhang, Fengwu Wang

Proceedings Volume 12081, Sixth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2021); 1208124 (2022) https://doi.org/10.1117/12.2623859
Event: Sixth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2021), 2021, Chongqing, China

Abstract

Traffic flow prediction is one of the point research contents of transportation engineering. Accurate ship traffic flow prediction is the value to ensure the safety of ship navigation and smooth channel. In order to predict the port ship traffic flow more accurately, aiming at the shortcomings of the traditional prediction model, a port ship traffic flow prediction model based on long-term and short-term memory network (LSTM) is proposed. Finally, Qingdao port is taken as an example to predict and compare with ARIMA model. The results show that, compared with ARIMA model, the mean absolute percentage error (MAPE) of LSTM model is as low as 3.476%, which indicates that the prediction accuracy of LSTM model is higher and it can be well applied to the field of ship traffic flow prediction. According to the prediction results, it can provide basic basis for channel planning and design and ship navigation management, maximize the navigation capacity of the channel, and optimize the allocation of port resources.

Citation Download Citation

Zhe Ji, Le Wang, Xiaobo Zhang, and Fengwu Wang "Ship traffic flow forecast of Qingdao port based on LSTM", Proc. SPIE 12081, Sixth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2021), 1208124 (6 February 2022); https://doi.org/10.1117/12.2623859

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available