Natural optimization algorithms for the cross-dock door assignment problem

Abstract

Cross-docking is a practice in logistics in which shipments are directly moved from an inbound truck into an outbound truck. A recognized problem in this domain is the assignment of trucks to doors in a way that the distance to be traveled between the doors is minimized. This problem is known as the cross-dock door assignment problem (CDAP). The purpose of this paper is to present a novel algorithm that minimizes the traveling distance of the handling machines when moving cargo from an inbound truck to an outbound truck. A lot of research has been conducted regarding this topic; still, up to our knowledge, none used scatter search (SS). This paper modifies a classical mathematical model that represents the CDAP and implements an evolutionary metaheuristic SS algorithm and tests it and then compares the results with those of another evolutionary algorithm, i.e., genetic algorithm (GA). The results indicate that the SS algorithm outperformed the GA, particularly for large-sized problems with a diverse reference set.