Traditionally, terminal operators create an initial berthing plan before the arrival of incoming vessels. This plan involves decisions on when and where to load or discharge containers for the calling vessels. However, disruptive unforeseen events (i.e., arrival delays, equipment breakdowns, tides, or extreme weather) interfere with the implementation of this initial plan. For terminals, berths and quay cranes are both crucial resources, and their capacity limits the efficiency of port operations. Thus, one way to minimize the adverse effects caused by disruption is to ally different terminals to share berthing resources. In some challenging situations, terminal operators also need to consider the extensive transshipment connections between feeder and mother vessels. Therefore, in this work, we investigate a collaborative variant of the berth allocation recovery problem which focuses on the collaboration among terminals and transshipment connections between vessels. We propose a mixed-integer programming model to (re)-optimize the initial berth and quay crane allocation plan and develop a Squeaky Wheel Optimization metaheuristic to find near-optimal solutions for large-scale instances. The results from the performed computational experiments, considering multiple scenarios with disruptive events, show consistent improvements of up to 40% for the suggested collaborative strategy (in costs for the terminal operators).

Berth allocation is fundamental to port-related operations in maritime shipping. Port managers have to deal with the increasing demands either by expanding the terminals or by improving efficiency to maintain competitiveness. Port expansion is a long-term project, and it requires much capital investment. Thus, the question of how to enhance the efficiency of berth allocation has received much research interest. Research on the Berth Allocation Problem (BAP) in container ports is quite advanced. However, only limited research focuses on BAP in bulk ports, although some similarities exist. Contributing to Operations Research approaches on the BAP, this paper develops a hybrid BAP mixed-integer optimization model dedicated to bulk ports. In addition to considering the handling characteristics of bulk ports, we also incorporate more practical factors such as unavailability and stock levels. The objective of the proposed model is to minimize the demurrage fee for all vessels under consideration of unavailability and stock constraints. We use the commercial software CPLEX to obtain the optimal solutions for a set of distinct instances, explicitly considering the situation of multiple cargo types on one vessel, which provides a better fit for the loading or discharging operations in real-world bulk ports. This is the first study to our knowledge that dedicates itself to the BAP in bulk ports and considers unavailability and stock constraints simultaneously. Our solutions can provide timely and effective decision support to bulk port managers.