Integer programming model and branch-and-cut algorithm for the stack inbound and pre-marshalling problem

Abstract

The steel plate yard of shipbuilding separates the inbound and pre-processing operations, increasing the number of moves. This paper considers the mixed inbound and pre-processing operation and studies the stack inbound and pre-marshalling problem where storage and relocation moves can alternate. This problem aims to find a minimum operation to store all inbound plates while eliminating all blocking plates. We propose a novel integer programming model combining two moves in one time period. This model reduces the number of periods, thereby decreasing the model size, but requires extra constraints to avoid infeasible patterns. An exact branch-and-cut algorithm is introduced to tackle the influence of these extra constraints. This paper provides a new modeling approach for stack-related problems, and the experiments show that the proposed method outperforms other ILP-based methods in the literature.