Machine-Learning-Based Prediction of Port-Channel Traffic Demand and Capacity Adaptability Assessment - A Case Study of the Core Port Area of Ningbo-Zhoushan Port
DOI:
https://doi.org/10.70088/a4sask75Keywords:
port channel, automatic identification system, machine learning, traffic demand forecasting, navigational capacity, capacity adaptabilityAbstract
The sustained growth of port throughput and the increasing size of vessels have rendered the relationship between channel traffic demand and navigational capacity progressively complex. Assessing channel development requirements solely on the basis of cargo throughput growth fails to capture shifts in vessel-type composition, temporal traffic clustering, or the disproportionate channel resources consumed by large vessels. Taking the core port area of Ningbo-Zhoushan Port as the study area, this paper investigates port-channel traffic-demand forecasting and capacity adaptability by integrating port statistics, publicly available Automatic Identification System (AIS) data, and channel-operation records. First, changes in port transport demand from 2015 to 2025 are analyzed, and hourly vessel traffic flows are reconstructed from AIS records. Second, a short-term traffic-demand forecasting framework comprising Autoregressive Integrated Moving Average (ARIMA), Support Vector Regression (SVR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) is established. On this basis, differences in channel-resource occupation among vessel types are considered, actual vessel calls are converted into equivalent traffic demand, and channel operating conditions are evaluated using the volume-to-capacity ratio. The results indicate that cargo throughput at Ningbo-Zhoushan Port increased from 890 million tonnes in 2015 to 1.432 billion tonnes in 2025, with a compound annual growth rate of approximately 4.87%. Growth in port throughput and vessel calls does not occur proportionally; vessel upsizing exerts a dual effect of reducing transport vessel calls while increasing the channel resources occupied per vessel. Annual traffic flow through the Xiazhimen Channel is approximately 50,000 vessel transits and is already approaching saturation. Following the commissioning of the 300,000-dwt Tiaozhoumen Channel at the end of 2025, the core port area formed two primary 300,000-dwt channels, and the passage capacity for ultra-large vessels is expected to increase by more than 50%. The study concludes that port-channel planning should transition from throughput extrapolation toward AIS-based, vessel-type-specific traffic-demand forecasting, with particular attention to changes in the peak-period volume-to-capacity ratio.References
H. Li, H. Jiao, and Z. Yang, "Ship trajectory prediction based on machine learning and deep learning: A systematic review and methods analysis," Engineering Applications of Artificial Intelligence, vol. 126, p. 107062, 2023.
H. Li, H. Jiao, and Z. Yang, "AIS data-driven ship trajectory prediction modelling and analysis based on machine learning and deep learning methods," Transportation Research Part E: Logistics and Transportation Review, vol. 175, p. 103152, 2023.
Z. Liu, W. Chen, C. Liu, R. Yan, and M. Zhang, "A data mining-then-predict method for proactive maritime traffic management by machine learning," Engineering Applications of Artificial Intelligence, vol. 135, p. 108696, 2024.
H. Yu, X. Cui, X. Bai, C. Chen, and L. Xu, "Incorporating graph theory and time series analysis for fine-grained traffic flow prediction in port areas," Ocean Engineering, vol. 335, p. 121693, 2025.
L. Cazzanti, A. Davoli, and L. M. Millefiori, "Automated port traffic statistics: From raw data to visualisation," in 2016 IEEE International Conference on Big Data (Big Data), Dec. 2016, pp. 1569–1573.
K. N. Mitchell and B. Scully, "Waterway performance monitoring with automatic identification system data," Transportation Research Record, vol. 2426, no. 1, pp. 20–26, 2014.
B. M. Scully and K. F. Chambers, "Measuring port disruptions with automatic identification system data," in 15th Triennial International Conference, Reston, VA: American Society of Civil Engineers, Sep. 2019, pp. 311–321.
S. L. Chao and Y. L. Lin, "Gate automation system evaluation: a case of a container number recognition system in port terminals," Maritime Business Review, vol. 2, no. 1, pp. 21–35, 2017.
G. Pallotta, M. Vespe, and K. Bryan, "Vessel pattern knowledge discovery from AIS data: A framework for anomaly detection and route prediction," Entropy, vol. 15, no. 6, pp. 2218–2245, 2013.
G. C. Kessler, J. P. Craiger, and J. C. Haass, "A taxonomy framework for maritime cybersecurity: A demonstration using the automatic identification system," *TransNav: International Journal on Marine Navigation and Safety of Sea Transportation*, vol. 12, no. 3, pp. 429–437, 2018.
A. Hadjipieris, N. Dimitriou, and O. Arandjelović, "Unsupervised port berth localization from automatic identification system data," Sensors, vol. 25, no. 22, p. 6845, 2025.
A. Harati-Mokhtari, A. Wall, P. Brooks, and J. Wang, "Automatic Identification System (AIS): Data reliability and human error implications," The Journal of Navigation, vol. 60, no. 3, pp. 373–389, 2007.
O. M. Belcore, M. Di Gangi, and A. Polimeni, "Automatic identification system data to assess container port performances," Transportation Research Procedia, vol. 86, pp. 612–619, 2025.
D. McFarlane and Y. Sheffi, "The impact of automatic identification on supply chain operations," The International Journal of Logistics Management, vol. 14, no. 1, pp. 1–17, 2003.
T. D. Nguyen, S. Marchal, M. Miettinen, H. Fereidooni, N. Asokan, and A. R. Sadeghi, "DÏoT: A federated self-learning anomaly detection system for IoT," in 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS), Jul. 2019, pp. 756–767.
S. Brusch, S. Lehner, T. Fritz, M. Soccorsi, A. Soloviev, and B. Van Schie, "Ship surveillance with TerraSAR-X," IEEE Transactions on Geoscience and Remote Sensing, vol. 49, no. 3, pp. 1092–1103, 2010.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Qiang Tan (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.









