Reinforcement Learning Driven Multi-Agent Prediction of Ship Fuel Consumption Across Sea States

Abstract

Accurate prediction of ship fuel consumption under complex sea conditions is of great significance for promoting green shipping. However, achieving consistently high accuracy is challenging due to the dynamic nature of maritime environments, where static prediction models often fail to adapt effectively. This study proposed a multi-agent prediction method for predicting ship fuel consumption, using reinforcement learning to dynamically select the optimal prediction model to adapt to changes in sea state. The proposed method leverages Proximal Policy Optimization (PPO) to guide the selection process, enabling the system to dynamically adapt to environmental variations. Experiments based on real-world seagoing ship data demonstrate that the PPO-based strategy consistently achieves high predictive accuracy, maintaining R² above 0.89 across all sea states and achieving substantial error reductions compared with conventional single-model approaches. The results highlight the method’s engineering significance in achieving adaptive and energy-efficient ship operation under uncertain sea conditions.

Keywords

Ship fuel consumption prediction / Multi-agent learning / Reinforcement learning / Energy efficiency / Maritime energy modeling

References

[1] Alazemi F, Alazmi A, Alrumaidhi M, Molden N. Predicting fuel consumption and emissions using GPS-based machine learning models for gasoline and diesel vehicles. Sustainability, 2025, 17(6): 2395-2413

[2] Calvaresi D, Dicente Cid Y, Marinoni M, Dragoni AF, Najjar A, Schumacher M (2021) Real-time multi-agent systems: rationality, formal model, and empirical results. Autonomous Agents and Multi-Agent Systems 35(1). https://doi.org/10.1007/s10458-020-09492-5

[3] Chuah LF, Mokhtar K, Mhd Ruslan SM, Bakar AA, Abdullah MA, Osman NH, Bokhari A, Mubashir M, Show PL. Implementation of the energy efficiency existing ship index and carbon intensity indicator on domestic ship for marine environmental protection. Environ Res, 2023, 222: 115348-115358

[4] Fan A, Yang J, Yang L, Wu D, Vladimir N. A review of ship fuel consumption models. Ocean Engineering, 2022, 264: 112405-112422

[5] Hu Z, Fan A, Mao W, Shu Y, Wang Y, Xia M, Yi Q, Li B. Ship energy consumption prediction: Multi-model fusion methods and multi-dimensional performance evaluation. Ocean Engineering, 2025, 322: 120538-120555

[6] Hu Z, Zhou T, Zhen R, Jin Y, Li X, Osman MT. A two-step strategy for fuel consumption prediction and optimization of ocean-going ships. Ocean Engineering, 2022, 249: 110904-110919

[7] Huang L, Pena B, Liu Y, Anderlini E. Machine learning in sustainable ship design and operation: A review. Ocean Engineering, 2022, 266: 112907-112925

[8] Ilias L, Kapsalis P, Mouzakitis S, Askounis D. A multitask learning framework for predicting ship fuel oil consumption. IEEE Access, 2023, 11: 132576-132589

[9] Işıklı E, Aydın N, Bilgili L, Toprak A. Estimating fuel consumption in maritime transport. Journal of Cleaner Production, 2020, 275: 124142-124153

[10] Islam Rony Z, Mofijur M, Hasan MM, Rasul MG, Jahirul MI, Forruque Ahmed S, Kalam MA, Anjum Badruddin I, Yunus Khan TM, Show P-L. Alternative fuels to reduce greenhouse gas emissions from marine transport and promote UN sustainable development goals. Fuel, 2023, 338: 127220-127232

[11] Kim D-K, Kim K. A convolutional transformer model for multivariate time series prediction. IEEE Access, 2022, 10: 101319-101329

[12] Lang X, Wu D, Mao W. Physics-informed machine learning models for ship speed prediction. Expert Systems with Applications, 2024, 238: 121877-121893

[13] Li T, Zhu K, Luong NC, Niyato D, Wu Q, Zhang Y, Chen B. Applications of multi-agent reinforcement learning in future internet: a comprehensive survey. IEEE Communications Surveys & Tutorials, 2022, 24(2): 1240-1279

[14] Liu L, Zhang Y, Hu Y, Wang Y, Sun J, Dong X. A hybrid-clustering model of ship trajectories for maritime traffic patterns analysis in port area. Journal of Marine Science and Engineering, 2022, 10(3): 342-369

[15] Ma L, Yang P, Gao D, Bao C. A multi-objective energy efficiency optimization method of ship under different sea conditions. Ocean Engineering, 2023, 290: 116337-116347

[16] Maldonado D, Cruz E, Abad Torres J, Cruz PJ, Gamboa Benitez S. Multi-agent systems: a survey about its components, framework and workflow. IEEE Access, 2024, 12: 80950-80975

[17] Mequanenit AM, Nibret EA, Herrero-Martín P, García-González MS, Martínez-Béjar R. A multi-agent deep reinforcement learning system for governmental interoperability. Applied Sciences, 2025, 15(6): 3146-3169

[18] Mi JJ, Wang Y, Zhang N, Zhang C, Ge J. A bibliometric analysis of green shipping: research progress and challenges for sustainable maritime transport. Journal of Marine Science and Engineering, 2024, 12(10): 1787-1812

[19] Peng Y, Liu H, Li X, Huang J, Wang W. Machine learning method for energy consumption prediction of ships in port considering green ports. Journal of Cleaner Production, 2020, 264: 121564-121578

[20] Petersen JP, Jacobsen DJ, Winther O. Statistical modelling for ship propulsion efficiency. Journal of Marine Science and Technology, 2011, 17(1): 30-39

[21] Rio A, Jimenez D, Serrano J. Comparative analysis of A3C and PPO algorithms in reinforcement learning: a survey on general environments. IEEE Access, 2024, 12: 146795-146806

[22] Shakya AK, Pillai G, Chakrabarty S. Reinforcement learning algorithms: A brief survey. Expert Systems with Applications, 2023, 231: 120495-120527

[23] Shi Z, Du L, He X, Gao X, Wu H, Liu Y, Ma H, Huo X, Chen X. Prediction model of yield strength of V–N steel hot-rolled plate based on machine learning algorithm. The Journal of The Minerals, 2023, 75(5): 1750-1762

[24] Song T, Pang C, Hou B, Xu G, Xue J, Sun H, Meng F (2023) A review of artificial intelligence in marine science. Frontiers in Earth Science 11. https://doi.org/10.3389/feart.2023.1090185

[25] Vorkapić A, Radonja R, Martinčić-Ipšić S. Predicting seagoing ship energy efficiency from the operational data. Sensors, 2021, 21(8): 2832-2851

[26] Wang G, Wei F, Jiang Y, Zhao M, Wang K, Qi H. A multi-AUV maritime target search method for moving and invisible objects based on multi-agent deep reinforcement learning. Sensors (Basel), 2022, 22(21): 8562-8574

[27] Wang H, Wu Y, Min G, Xu J, Tan P. Data-driven dynamic resource scheduling for network slicing: a deep reinforcement learning approach. Information Sciences, 2019, 498: 106-116

[28] Wang H, Yan R, Wang S, Zhen L. Innovative approaches to addressing the tradeoff between interpretability and accuracy in ship fuel consumption prediction. Transportation Research Part C: Emerging Technologies, 2023, 157: 104361-104377

[29] Wang J, Bielicki S, Kluwe F, Orihara H, Xin G, Kume K, Oh S, Liu S, Feng P. Validation study on a new semi-empirical method for the prediction of added resistance in waves of arbitrary heading in analyzing ship speed trial results. Ocean Engineering, 2021, 240: 109959-109975

[30] Wang K, Liu X, Guo X, Wang J, Wang Z, Huang L. A novel high-precision and self-adaptive prediction method for ship energy consumption based on the multi-model fusion approach. Energy, 2024, 310: 1033265-1033281

[31] Wang K, Yan X, Yuan Y, Jiang X, Lin X, Negenborn RR. Dynamic optimization of ship energy efficiency considering time-varying environmental factors. Transportation Research Part D: Transport and Environment, 2018, 62: 685-698

[32] Wang Y, Zhao Y. Multiple ships cooperative navigation and collision avoidance using multi-agent reinforcement learning with communication. Ocean Engineering, 2025, 320: 120244-120261

[33] Xia Z, Guo Z, Wang W, Jiang Y. Joint optimization of ship scheduling and speed reduction: A new strategy considering high transport efficiency and low carbon of ships in port. Ocean Engineering, 2021, 233: 109224-109237

[34] Xie X, Sun B, Li X, Olsson T, Maleki N, Ahlgren F. Fuel consumption prediction models based on machine learning and mathematical methods. Journal of Marine Science and Engineering, 2023, 11(4): 738-757

[35] Xue J, Yang P, Li Q, Song Y, Gelder PHAJM, Papadimitriou E, Hu H. Machine learning in maritime safety for autonomous shipping: a bibliometric review and future trends. Journal of Marine Science and Engineering, 2025, 13(4): 746-794

[36] Yang L, Chen G, Zhao J, Rytter NGM. Ship speed optimization considering ocean currents to enhance environmental sustainability in maritime shipping. Sustainability, 2020, 12(9): 3649-3673

[37] Yang Z, Qu W, Zhuo J. Optimization of energy consumption in ship propulsion control under severe sea conditions. Journal of Marine Science and Engineering, 2024, 12(9): 1461-1478

[38] Zhou T, Hu Q, Hu Z, Zhen R. An adaptive hyper parameter tuning model for ship fuel consumption prediction under complex maritime environments. Journal of Ocean Engineering and Science, 2022, 7(3): 255-263

RIGHTS & PERMISSIONS

Harbin Engineering University and Springer-Verlag GmbH Germany, part of Springer Nature

Just Accepted

This article has successfully passed peer review and final editorial review, and will soon enter typesetting, proofreading and other publishing processes. The currently displayed version is the accepted final manuscript. The officially published version will be updated with format, DOI and citation information upon launch. We recommend that you pay attention to subsequent journal notifications and preferentially cite the officially published version. Thank you for your support and cooperation.