An NSABC algorithm for multi-aisle AS/RS scheduling optimization

Abstract

A multi-objective scheduling optimization model for multi-aisle AS/RS is established in this study, considering the sequence constraint of inbound and outbound tasks with a multi-carrier S/R machine. The aisle switching time is redefined according to the most common layout, which is more realistic than in some of the literature. A non-dominated sorting artificial bee colony (NSABC) algorithm is proposed to solve the problem, which combines the multi-objective handling framework of NSGA-II and the neighborhood search of the ABC algorithm. The encoding structure, decoding method, and updating operators of the solutions are redesigned according to the features of the model. In experiments, operators and parameters of the NSABC algorithm are tested and fine-tuned. The best version is then tested on several groups of instances and compared with the EMBBO, NSGA-II, and MOEA/D algorithms. The results show that the proposed NSABC algorithm outperforms the comparison algorithms in most instances, which indicates that it is an effective approach for the AS/RS scheduling optimization problem.