Energy-efficient tugboat scheduling: A hybrid transformer-attention mechanism and artificial multiple intelligence system

Abstract

Tugboats play a crucial role in connecting maritime and inland logistics by transferring goods from large vessels. However, managing their energy consumption is a major challenge due to factors such as barge capacity, cargo weight, tidal schedules, navigational complexities, and regulatory constraints. Efficient scheduling is essential to minimizing costs and enhancing sustainability. To address this challenge, this study introduces a mixed-integer programming model to optimize tugboat scheduling, incorporating real-world constraints to reduce energy consumption and operational inefficiencies. To address industrial scale problems, we propose a Hybrid Transformer-Attention Mechanism and Artificial Multiple Intelligence System (HT-AMIS), combined with metaheuristic-inspired intelligence boxes (IBs), to enhance adaptability and efficiency. The results show that HT-AMIS reduces tugboat operating costs by 11.75%, with energy costs reduced by 10.73% and penalty costs reduced by 21.96%. The model demonstrated robustness, effectively handling challenging scenarios such as tugboat breakdowns and severe weather conditions.