Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches

• Autorzy: Pham Nguyen Dang Khoa , Dinh Gia Huy , Nguyen Canh Lam , Dang Hai Quoc , Pham Hoang Thai , Nguyen Quyen Tat , Tran Minh Cong

Identyfikatory

DOI

10.2478/pomr-2025-0008

• Języki publikacji: EN

Abstrakty

EN

Rising shipping emissions greatly affect greenhouse gas (GHG) levels, so precise fuel consumption forecasting is essential to reduce environmental effects. Precision forecasts using machine learning (ML) could offer sophisticated solutions that increase the fuel efficiency and lower emissions. Indeed, five ML techniques, linear regression (LR), decision tree (DT), random forest (RF), XGBoost, and AdaBoost, were used to develop ship fuel consumption models in this study. It was found that, with an R² of 1, zero mean squared error (MSE), and a negligible mean absolute percentage error (MAPE), the DT model suited the training set perfectly, while R² was 0.8657, the MSE was 56.80, and the MAPE was 16.37% for the DT model testing. More importantly, this study provided Taylor diagrams and violin plots that helped in the identification of the best-performing models. Generally, the employed ML approaches efficiently predicted the data; however, they are black-box methods. Hence, explainable machine learning methods like Shapley additive explanations, the DT structure, and local interpretable model-agnostic explanations (LIME) were employed to comprehend the models and perform feature analysis. LIME offered insights, demonstrating that the major variables impacting predictions were distance (≤450.88 nm) and time (40.70 < hr ≤ 58.05). By stressing the most important aspects, LIME can help one to comprehend the models with ease.

Słowa kluczowe

EN

ship fuel consumption; fuel consumption forecasting; model prediction; machine learning technique; operation efficiency

• Wydawca: Gdańsk University of Technology, Faculty of Ocean Engineering & Ship Technology
• Czasopismo: Polish Maritime Research
• Rocznik: 2025
• Tom: nr 1
• Strony: 81--94
• Opis fizyczny: Bibliogr. 55 poz., rys., tab.

Twórcy

autor

Pham Nguyen Dang Khoa

• Ho Chi Minh City University of Transport, Ho Chi Minh city, Viet Nam

• https://orcid.org/0000-0003-1750-2077

autor

Dinh Gia Huy

• Ho Chi Minh City University of Transport, Ho Chi Minh city, Viet Nam

• https://orcid.org/0000-0001-7835-1132

autor

Nguyen Canh Lam

• The Business School, Business Innovation Department, RMIT University, Ho Chi Minh city, Viet Nam

autor

Dang Hai Quoc

• Ho Chi Minh City University of Transport, Ho Chi Minh city, Viet Nam

• https://orcid.org/0009-0007-4333-1378

autor

Pham Hoang Thai

• Ho Chi Minh City University of Transport, Ho Chi Minh city, Viet Nam

• https://orcid.org/0009-0005-5806-7181

autor

Nguyen Quyen Tat

• Ho Chi Minh City University of Transport, Ho Chi Minh city, Viet Nam

autor

Tran Minh Cong

• Nha Trang University, Nha Trang, Viet Nam

Bibliografia

• 1 Nguyen HP, Nguyen CTU, Tran TM, Dang QH, Pham NDK. Artificial intelligence and machine learning for green shipping: Navigating towards sustainable maritime practices. JOIV Int J Informatics Vis 2024;8:1. https://doi.org/10.62527/joiv.8.1.2581.
• 2 Duc BV, Nguyen HP. A comprehensive review on big data-based potential applications in marine shipping management. Int J Adv Sci Eng Inf Technol 2021;11:1067–77.
• 3 UNCTAD. Review of Maritime Report 2021. 2021.
• 4 Zeńczak W, Gromadzińska AK. Preliminary analysis of the use of solid biofuels in a ship’s power system. Polish Marit Res 2020;27:67–79. https://doi.org/10.2478/pomr-2020-0067.
• 5 Hoang AT, Foley AM, Nižetić S, Huang Z, Ong HC, Ölçer AI, et al. Energy-related approach for reduction of CO2 emissions: A critical strategy on the port-to-ship pathway. J Clean Prod 2022;355:131772. https://doi.org/10.1016/j.jclepro.2022.131772.
• 6 Vakili S, Ölçer AI, Schönborn A, Ballini F, Hoang AT. Energy‐related clean and green framework for shipbuilding community towards zero‐emissions: A strategic analysis from concept to case study. Int J Energy Res 2022;46:20624–49. https://doi.org/10.1002/er.7649.
• 7 Su W, Zhang Q, Liu Y. Event-triggered adaptive neural network trajectory tracking control for underactuated ships under uncertain disturbance. Polish Marit Res 2023;30:119–31. https://doi.org/doi:10.2478/pomr-2023-0045.
• 8 Sahnen D, Harmsen J, Nesterova N, Bekdemir C, Van Kranenburg K, Andersson K, et al. Reducing sulphur emissions from ships - The impact of international regulation. Int J Hydrogen Energy 2020;12:32–5.
• 9 Le TT, Sharma P, Pham NDK, Le DTN, Le VV, Osman SM, et al. Development of comprehensive models for precise prognostics of ship fuel consumption. J Mar Eng Technol 2024:1–15. https://doi.org/10.1080/20464177.2024.2372888.
• 10 Hoang AT, Tran VD, Dong VH, Le AT. An experimental analysis on physical properties and spray characteristics of an ultrasound-assisted emulsion of ultra-low-sulphur diesel and Jatropha-based biodiesel. J Mar Eng Technol 2022;21:73–81. https://doi.org/10.1080/20464177.2019.1595355.
• 11 Nguyen VN, Chung N, Balaji GN, Rudzki K, Hoang AT. Internet of things-driven approach integrated with explainable machine learning models for ship fuel consumption prediction. Alexandria Eng J 2025;118:664- 80. https://doi.org/10.1016/j.aej.2025.01.067.
• 12 Nguyen HP, Hoang AT, Nizetic S, Nguyen XP, Le AT, Luong CN, et al. The electric propulsion system as a green solution for management strategy of CO2 emission in ocean shipping: A comprehensive review. Int Trans Electr Energy Syst 2021;31:e12580. https://doi.org/10.1002/2050-7038.12580.
• 13 Mallouppas G, Yfantis EA. Decarbonization in shipping industry: A review of research, technology development, and innovation proposals. J Mar Sci Eng 2021;9:415. https://doi.org/10.3390/jmse9040415.
• 14 Herdzik J. Decarbonization of marine fuels—The future of shipping. Energies 2021;14:4311. https://doi.org/10.3390/en14144311.
• 15 Hoang AT, Pandey A, Martinez De Osés FJ, Chen W-H, Said Z, Ng KH, et al. Technological solutions for boosting hydrogen role in decarbonization strategies and net-zero goals of world shipping: Challenges and perspectives. Renew Sustain Energy Rev 2023;188:113790. https://doi.org/10.1016/j.rser.2023.113790.
• 16 IMO. Fourth IMO GHG Study 2020 full report. Int Marit Organ 2021;6:524.
• 17 Chen Y, Sun B, Xie X, Li X, Li Y, Zhao Y. Short-term forecasting for ship fuel consumption based on deep learning. Ocean Eng 2024;301:117398. https://doi.org/10.1016/j.oceaneng.2024.117398.
• 18 Mantoju CD. Analysis of MARPOL implementation based on port state control statistics. J Int Marit Safety, Environ Aff Shipp 2021;5:132–45. https://doi.org/10.1080/25725084.2021.1965281.
• 19 Schinas O, Stefanakos CN. Selecting technologies towards compliance with MARPOL Annex VI: The perspective of operators. Transp Res Part D Transp Environ 2014;28:28–40. https://doi.org/10.1016/j.trd.2013.12.006.
• 20 Hussain N, Khan A, Shumaila, Memon S. Addressing marine pollution: An analysis of MARPOL 73/78 regulations and global implementation efforts. J Soc Sci Rev 2023;3:572–89. https://doi.org/10.54183/jssr.v3i1.193.
• 21 Čampara L, Hasanspahić N, Vujičić S. Overview of MARPOL ANNEX VI regulations for prevention of air pollution from marine diesel engines. SHS Web Conf 2018;58:01004. https://doi.org/10.1051/shsconf/20185801004.
• 22 MEPC. MEPC 67 October 2014.
• 23 Chircop A. The IMO initial strategy for the reduction of GHGs from international shipping: A commentary. Int J Mar Coast Law 2019;34:482–512. https://doi.org/10.1163/15718085-13431093.
• 24 Joung T-H, Kang S-G, Lee J-K, Ahn J. The IMO initial strategy for reducing greenhouse gas (GHG) emissions, and its follow-up actions towards 2050. J Int Marit Safety, Environ Aff Shipp 2020;4:1–7. https://doi.org/10.1080/25725084.2019.1707938.
• 25 Nguyen VN, Rudzki K, Dzida M, Pham NDK, Pham MT, Nguyen PQP, et al. Understanding fuel saving and clean fuel strategies towards green maritime. Polish Marit Res 2023;30:146–64. https://doi.org/10.2478/pomr-2023-0030.
• 26 Ghaemi MH, Zeraatgar H. Impact of propeller emergence on hull, propeller, engine, and fuel consumption performance in regular head waves. Polish Marit Res 2022;29:56–76. https://doi.org/10.2478/pomr-2022-0044.
• 27 Korczewski Z. Energy and emission quality ranking of newly produced low-sulphur marine fuels. Polish Marit Res 2022;29:77–87. https://doi.org/10.2478/pomr-2022-0045.
• 28 Zhang M, Tsoulakos N, Kujala P, Hirdaris S. A deep learning method for the prediction of ship fuel consumption in real operational conditions. Eng Appl Artif Intell 2024;130:107425. https://doi.org/10.1016/j.engappai.2023.107425.
• 29 Agand P, Kennedy A, Harris T, Bae C, Chen M, Park EJ. Fuel consumption prediction for a passenger ferry using machine learning and in-service data: A comparative study. Ocean Eng 2023;284:115271. https://doi.org/10.1016/j.oceaneng.2023.115271.
• 30 Chen ZS, Lam JSL, Xiao Z. Prediction of harbour vessel fuel consumption based on machine learning approach. Ocean Eng 2023;278:114483. https://doi.org/10.1016/j.oceaneng.2023.114483.
• 31 Yan R, Wang S, Du Y. Development of a two-stage ship fuel consumption prediction and reduction model for a dry bulk ship. Transp Res Part E Logist Transp Rev 2020;138:101930. https://doi.org/10.1016/j.tre.2020.101930.
• 32 Rudzki K, Gomulka P, Hoang AT. Optimization model to manage ship fuel consumption and navigation time. Polish Marit Res 2022;29:141–53. https://doi.org/10.2478/pomr-2022-0034.
• 33 Shanmugasundar G, Vanitha M, Čep R, Kumar V, Kalita K, Ramachandran M. A comparative study of linear, random forest and AdaBoost regressions for modeling non-traditional machining. Processes 2021;9:2015. https://doi.org/10.3390/pr9112015.
• 34 Quirk TJ. Correlation and simple linear regression. Excel 2016 Appl. Stat. High Sch. Students, Cham: Springer International Publishing; 2018, pp. 107–52. https://doi.org/10.1007/978-3-319-89993-0_6.
• 35 Breiman L. Random forests. Mach Learn 2001;45:5–32. https://doi.org/10.1023/A:1010933404324.
• 36 Gnecco N, Terefe EM, Engelke S. Extremal random forests. J Am Stat Assoc 2024:1–24. https://doi.org/10.1080/01621459.2023.2300522.
• 37 Speiser JL, Miller ME, Tooze J, Ip E. A comparison of random forest variable selection methods for classification prediction modeling. Expert Syst Appl 2019;134:93–101. https://doi.org/10.1016/j.eswa.2019.05.028.
• 38 Sprenger M, Schemm S, Oechslin R, Jenkner J. Nowcasting Foehn wind events using the AdaBoost machine learning algorithm. Weather Forecast 2017;32:1079–99. https://doi.org/10.1175/WAF-D-16-0208.1.
• 39 Kyeremeh F, Zhi F, Yi Y, Gyamfi E, Nti IK. Solar PV power forecasting with a hybrid LSTM-AdaBoost ensemble. 2022 IEEE/IET Int. Util. Conf. Expo., IEEE; 2022, pp. 1–7. https://doi.org/10.1109/IUCE55902.2022.10079424.
• 40 Chen T, Guestrin C. XGBoost. Proc. 22nd ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., New York, NY, USA: ACM; 2016, pp. 785–94. https://doi.org/10.1145/2939672.2939785.
• 41 Darmawan H, Yuliana M, Hadi MZS. GRU and XGBoost performance with hyperparameter tuning using GridSearchCV and Bayesian optimization on an IoT-based weather prediction system. Int J Adv Sci Eng Inf Technol 2023;13:848–59. https://doi.org/10.18517/ijaseit.13.3.18377.
• 42 Abdi J, Hadavimoghaddam F, Hadipoor M, Hemmati-Sarapardeh A. Modeling of CO2 adsorption capacity by porous metal organic frameworks using advanced decision tree-based models. Sci Rep 2021;11:24468. https://doi.org/10.1038/s41598-021-04168-w.
• 43 Sagi O, Rokach L. Approximating XGBoost with an interpretable decision tree. Inf Sci (Ny) 2021;572:522–42. https://doi.org/10.1016/j.ins.2021.05.055.
• 44 Jumin E, Basaruddin FB, Yusoff YBM, Latif SD, Ahmed AN. Solar radiation prediction using boosted decision tree regression model: A case study in Malaysia. Environ Sci Pollut Res 2021;28:26571–83. https://doi.org/10.1007/s11356-021-12435-6.
• 45 Sharma KV, Talpa Sai PHVS, Sharma P, Kanti PK, Bhramara P, Akilu S. Prognostic modeling of polydisperse SiO2/Aqueous glycerol nanofluids’ thermophysical profile using an explainable artificial intelligence (XAI) approach. Eng Appl Artif Intell 2023;126:106967. https://doi.org/10.1016/j.engappai.2023.106967.
• 46 Alruqi M, Sharma P, Algburi S, Khan MA, Alsubih M, Islam S. Biomass energy transformation: Harnessing the power of explainable AI to unlock the potential of ultimate analysis data. Environ Technol Innov 2024;35:103652. https://doi.org/10.1016/j.eti.2024.103652.
• 47 de Myttenaere A, Golden B, Le Grand B, Rossi F. Mean absolute percentage error for regression models. Neurocomputing 2016;192:38–48. https://doi.org/10.1016/j.neucom.2015.12.114.
• 48 Elvidge S, Angling MJ, Nava B. On the use of modified Taylor diagrams to compare ionospheric assimilation models. Radio Sci 2014;49:737–45. https://doi.org/10.1002/2014RS005435.
• 49 Simão ML, Videiro PM, Silva PBA, de Freitas Assad LP, Sagrilo LVS. Application of Taylor diagram in the evaluation of joint environmental distributions’ performances. Mar Syst Ocean Technol 2020;15:151–9. https://doi.org/10.1007/s40868-020-00081-5.
• 50 Aberasturi DT. Violin plot. Wiley StatsRef Stat. Ref. Online, Wiley; 2023, pp. 1–7. https://doi.org/10.1002/9781118445112.stat08426.
• 51 Molina E, Viale L, Vazquez P. How should we design violin plots? 2022 IEEE 4th Work. Vis. Guidel. Res. Des. Educ., IEEE; 2022, pp. 1–7. https://doi.org/10.1109/VisGuides57787.2022.00006.
• 52 Hintze JL, Nelson RD. Violin plots: A box plot-density trace synergism. Am Stat 1998;52:181. https://doi.org/10.2307/2685478.
• 53 Tanious R, Manolov R. Violin plots as visual tools in the meta-analysis of single-case experimental designs. Methodology 2022;18:221–38. https://doi.org/10.5964/meth.9209.
• 54 Vergura S. Supervision of the energy performance of multi-arrays PV plants by means of bi-monthly violin plots. Renew Energy Power Qual J 2022;20:625–8. https://doi.org/10.24084/repqj20.384.
• 55 Shin J, Yim I, Kwon S-B, Park S, Kim M, Cha Y. Evaluation of temperature effects on brake wear particles using clustered heatmaps. Environ Eng Res 2019;24:680–9. https://doi.org/10.4491/eer.2018.385.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki i promocja sportu (2025).