An intelligent hyperheuristic algorithm for the berth allocation and scheduling problem at marine container terminals

Berth allocation and scheduling at marine container terminals holds critical significance for optimizing port operations and ensuring efficient maritime logistics, yet it is exceptionally challenging due to its operational complexity and practical constraints. This study presents a mixed-integer linear programming model for the dynamic discrete berth allocation and scheduling problem with the objective to minimize the total turnaround cost and develop an efficient service schedule for each incoming vessel with a feasible service order and berthing position. A novel Hyperheuristic Hybridized with Exact Optimization (HHEO) is developed to explicitly solve the challenging decision problem studied herein. The HHEO algorithm is designed to dynamically select and apply different genetic operators based on their actual performance. Furthermore, intelligent exact optimization procedures specific to the domain of berth allocation and scheduling are periodically employed within the HHEO framework to improve solution quality and facilitate search for high-quality solutions. Detailed computational experiments are carried out to prove the superiority of the HHEO algorithm against solution methods inspired by exact optimization along with some well-known metaheuristic algorithms, demonstrating its efficiency and applicability for real-life berth planning at marine container terminals. Moreover, the experiments clearly demonstrate that incorporating a hyperheuristic framework along with problem-specific hybridization methods is critical for improving the exploratory and exploitative capabilities of the developed HHEO algorithm. Last but not least, valuable practical insights are uncovered through the proposed approach, offering reliable solutions that can support effective port management under different scenarios of berthing availability, vessel arrival intensity, and delayed vessel departure penalties.