The Multiport Berth Allocation Problem with Speed Optimization: Exact Methods and a Cooperative Game Analysis.

In this paper, we focus on a variant of the Berth Allocation Problem (BAP), which aims at assigning berthing times and positions to vessels in container terminals. The problem, known as the multi-port berth allocation problem (MPBAP) extends the BAP to cover multiple ports where vessel traveling speeds are optimized between ports, thus exploiting the potentials of a collaboration between carriers and terminal operators. Exploiting a graph representation of the problem, we reformulate an existing mixed-integer problem formulation into a generalized set partitioning problem where each variable refers to a sequence of feasible berths in the ports visited by the vessel. Integrating column generation and cut separation in a branch-and-cut-and-price procedure, the method is able to outperform commercial solvers in a set of benchmark instances and adapts better to larger instances. In addition, we apply methods of cooperative game theory for distributing efficiently the savings of a potential collaboration and show that both carriers and terminal operators would benefit from such collaboration.