Improved combinatorial optimization model for vessel scheduling in constrained channel ports

Vessel scheduling, which assigns time and space resources to vessels, is crucial for the safety of waterborne transportation and directly influences the efficiency of waterborne logistics systems. This paper proposes a systematic solution to the Multiple Legs and Multiple Navigation Modes Vessel Scheduling Problem (ML&MM-VSP) in the port region. First, a generalized framework for ML&MM-VSP is proposed, adaptable to different port layouts. Secondly, the original VSP model is simplified, and the constraint conditions are refined. The model sets total scheduling time and vessel delay time as two objectives while considering the characteristics of the waterway and vessel attributes as constraints. Thirdly, a Non-dominated Sorting Genetic Algorithm II (NSGA-II) is employed to solve the problem, aiming to identify optimal vessel scheduling strategies. Lastly, empirical analysis is conducted to verify the rationality and effectiveness of both the proposed model and the adopted algorithm, and comparative experiments are carried out to test ML&MM-VSP with different layouts. The results show that the proposed model performs better than the original models and can provide more effective guidance for ML&MM-VSP practice.