Objective Accurate prediction of ship fuel consumption is crucial for reducing operational costs and achieving energy-saving emission reduction. To this end, this study proposes a ship fuel consumption prediction model that integrates an improved grey wolf optimizer with the XGBoost algorithm, providing a reliable basis for ship energy efficiency management decisions.

Method This study develops an XGBoost-based fuel consumption prediction model using actual voyage data. To address the challenges posed by complex hyperparameter combinations and the inefficiency of manual tuning in the algorithm, an improved grey wolf optimizer is proposed for guided hyperparameter optimization of the XGBoost model. Building on the strengths of the traditional GWO, Tent Chaos mapping is introduced, along with horizontal and vertical crossover mechanisms and a boundary checking procedure during iterations, to train the XGBoost model with the optimal hyperparameter set.

Results Experimental results show that the predicted values of the ship fuel consumption model developed in this study closely align with the fitting line on the test set. The evaluation metrics (RMSE, and R²) surpass those of the untuned original XGBoost model and other mainstream models, demonstrating strong generalization ability and high prediction accuracy.

Conclusion The XGBoost model optimized via the improved grey wolf optimizer exhibits effective performance in predicting ship fuel consumption, providing reliable data support for energy-efficient operations and intelligent vessel navigation.