The CODeS research group is pleased to announce this seminar by Prof. Ramón Alvarez-Valdés, University of Valencia.

The stowage planning problem, in which a plan is developed for the position of the containers in a ship, is,  according to the state-of-the-art, most efficiently solved in two phases. In the first phase, Master Planning, groups of containers are assigned to locations in the ship. In the second phase, Slot Planning, the exact position of each container is determined.

In this talk,  we introduce first the stowage planning problem and then we focus on  the slot planning phase, in which the containers assigned to a container ship location have to be stowed, satisfying many conditions related to the way in which containers have to be stacked, the weight distribution, and the specific conditions regulating the containers with dangerous products.  The main objective is to pack as many containers as possible, all the assigned containers if they feasibly t into the location, and the secondary objective is to minimize the number of unproductive moves of containers that have to be removed just to get access to other containers below them.

In order to study the structure of the problem, we have developed an integer linear model.

In order to efficiently solve problems of realistic size, we have developed a GRASP algorithm which includes two constructive methods, several randomization strategies, and several improvement moves. The algorithms have been tested on both a set of real-world and custom generated instances. The results show that the GRASP algorithm obtains good results in a wide range of problems.

The stowage planning problem, in which a plan is developed for the position of the containers in a ship, is,  according to the state-of-the-art, most efficiently solved in two phases. In the first phase, Master Planning, groups of containers are assigned to locations in the ship. In the second phase, Slot Planning, the exact position of each container is determined.