Journal of Navigation and Port Research (한국항해항만학회지)

Korean Institute of Navigation and Port Research (한국항해항만학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on the Prediction of Fuel Consumption of Bulk Ship Main Engine Using Explainable Artificial Intelligence

SHAP을 활용한 벌크선 메인엔진 연료 소모량 예측연구

  • Hyun-Ju Kim (Intelligent Convergence Research Team, Korea Marine Equipment Research Institute) ;
  • Min-Gyu Park (Division of Industrial and Data Engineering, Pukyong National University) ;
  • Ji-Hwan Lee (Division of Industrial and Data Engineering, Pukyong National University)
  • 김현주 (한국조선해양기자재연구원) ;
  • 박민규 (부경대학교) ;
  • 이지환 (부경대학교 산업및데이터공학과)
  • Received : 2023.07.17
  • Accepted : 2023.08.04
  • Published : 2023.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

This study proposes a predictive model using XGBoost and SHapley Additive exPlanation (SHAP) to estimate fuel consumption in bulk carriers. Previous studies have also utilized ship engine data and weather data. However, they lacked reliability in predicted results and explanations of variables used in the fuel consumption prediction model implementation. To address these limitations, this study developed a predictive model using XGBoost and SHAP. It provides research background, scope, relevant regulations, previous studies, and research methodology. Additionally, it explains the data cleaning method for bulk carriers and verifies results of the predictive model.

본 연구에서는 벌크 선박의 연료 소비를 예측하기 위해 XGBoost와 SHapley Additive exPlanation (SHAP)을 사용하는 예측 모델을 제안한다. 기존 연구에서도 선박 엔진 데이터와 기상데이터를 활용하였지만 선박 연료소모량 예측 모델에 대한 예측 결과의 신뢰성과 예측 모델 구현에 사용된 변수들에 대한 설명이 부족한 한계가 있었다. 이러한 문제를 해결하기 위해 본 연구에서는 XGBoost와 SHAP를 사용하여 예측 모델을 개발하였다. 이 연구는 연구 배경, 범위, 관련 규정 및 이전 연구들, 그리고 연구 방법론에 대한 소개를 제공하며, 또한 벌크선 데이터 정제 방법과 예측 모델 결과의 검증을 설명한다.

Keywords

Acknowledgement

이 논문은 2023년도 해양수산부 재원으로 해양수산과학기술진흥원의 지원을 받아 수행된 연구이다. (20220469, 해양수산 내수시장활성화사업)

References

  1. Chen, T. and Guestrin, C.(2016), "XGBoost: A Scalable Tree Boosting System." Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785-794.
  2. Greenwell, B. M.(2017). "pdp:An R package for constructing partial dependence plots." The R Journal, Vol. 9, No. 1, pp. 421-436. https://doi.org/10.32614/RJ-2017-016
  3. Hannes, J. and Linda, S.(2015), "Increased energy efficiency in short sea shipping through decreased time in port", Transportation Research Part A: Policy and Practice, Vol. 71, pp. 167-178. https://doi.org/10.1016/j.tra.2014.11.008
  4. IMO(2023) "Amendments to MARPOL Annex VI and the NOx Technical Code 2008 with regard to the requirement for in-use NOx abatement technology", MEPC.377(80).
  5. Kim, Y. R., Kim, G. J. and Park, J. B.(2019), "A study on the Prediction of Fuel Consumption of a Ship Using the Principal Component Analysis", Journal of Korean Navigation and Port Research, Vol. 43, No. 6, pp. 335-343.
  6. Lloyd, S.(1953), "A Value for n-person Games" , Contributions to the Theory of Games II, pp. 307-317
  7. Lipton, Z. C.(2018), "The Mythos of Model Interpretability", Communications of the ACM, Vol. 61, No. 10, pp. 36-43. https://doi.org/10.1145/3233231
  8. Lundberg, S. M. and Lee, S. I.(2017), "A Unified Approach to Interpreting Model Predictions" Advances in Neural Information Processing Systems, Vol. 30, pp. 765-774.
  9. Moore, D. S. and Kirkland, S.(2007), "The Basic Practice of Statistics", Vol. 2, WH Freeman Press.
  10. Mengnan, D., Ninghao, L. and Xia, Hu.(2019), "Techniques for Interpretable Machine Learning" Communications of the ACM, Vol. 63, No. 1, pp. 68-77. https://doi.org/10.1145/3359786
  11. Opitz, D. and Maclin, R.(1999), "Popular Ensemble Methods: An Empirical Study." Journal of Artificial Intelligence Research, Vol. 11, pp. 169-198. https://doi.org/10.1613/jair.614
  12. Shin, D. W., Yang, C. S. and Jeon, H. G.(2021), "Real Ship Experimental Analysis on the Influence of Fuel Oil Consumption Considering Ship Speed and Ocean Environment." Journal of Coastal Disaster Prevention, Vol. 8, No. 2, pp. 99-110. https://doi.org/10.20481/kscdp.2021.8.2.99
  13. Stopford, M.(2009), "Maritime Economics. 3rd ed.", Oxford:Routledge.
  14. T, Uyanik., C, Karatug. and Y, Arslanoglu.(2020), "Machine Learning Approach to Ship Fuel Consumption: A Case of Container Vessel." Transportation Research Part D: Transport and Environment, Vol. 84, 102389.
  15. Tjoa, E. and Guan, C.(2020). "A survey on explainable artificial intelligence (xai): Toward medical xai." IEEE transactions on neural networks and learning systems, Vol. 2, No. 11, pp. 4793-4813.
  16. Yoon, Y. J.(2005), "Marine Navigation." Haean Publishing Company, pp. 1-20.