Reproduce “Data Fusion and Machine Learning for Ship Fuel Consumption Modelling - A Case of Bulk Carrier Vessel” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Abdella Mohamed, Xiangyu Hu, Christian Hendricks
Researchers from the Technical University of Munich and Everllence fused voyage reports with comprehensive hydrometeorological data from CMEMS and ERA5 to develop machine learning models for bulk carrier fuel consumption, achieving a predictive accuracy (R2) of 0.9021 with Random Forest and demonstrating the high influence of Engine RPM, temperature in surface snow, and seawater salinity.
Key insights
- Engine Rotation Per Minute (RPM) is a superior primary predictor for fuel consumption compared to Speed Over Ground (SOG), reflecting the engine's direct workload.
- Fusing extensive external hydrometeorological data with operational reports significantly enhances the accuracy of ship fuel consumption models, increasing R2 from approximately 0.76 to 0.90.
- Beyond Engine RPM, temperature in surface snow, seawater salinity, and propeller slip were identified as the most influential parameters for predicting fuel consumption.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.