Publication

Abstract : Estimating cargo power is vital because it allows for efficient planning and optimization of transportation logistics, ensuring the appropriate allocation of resources and maximizing operational effectiveness. It also facilitates the assessment and mitigation of the environmental impact of transportation by optimizing fuel consumption and reducing emissions. This chapter proposes a cutting-edge machine learning solution for accurate cargo power prediction. The Pearson correlation is employed to identify the most relevant features for the regression task, thereby aiding in the reduction of dimensionality. Multiple machine learning models are utilized for this purpose, and a comprehensive performance comparison is conducted. The results demonstrate that the proposed ensemble learning-based stacking model outperformed all other models, achieving an impressive coefficient of determination of 0.96. This research offers significant advancements to the field of shipping logistics optimization and provides practical insights for enhancing transportation efficiency and environmental sustainability.