Joint scheduling of parallel machines and AGVs with sequence-dependent setup times in a matrix workshop

Transportation systems are often ignored in traditional parallel-machine scheduling problems (PMSP). This study addresses the problem of joint scheduling of parallel machines and AGVs (PMSP-AGV) with sequence-dependent setup times in a matrix manufacturing workshop. The goal is to minimize the makespan from the departure of AGVs to completion of the last job. To achieve this goal, a mixed-integer linear programming model (MILP) is established. An efficient discrete artificial bee colony (DABC) algorithm was proposed with improvement strategies, including a two-stage heuristic for initializing the population, six effective neighborhood operators to explore neighborhood solutions, new algorithm control parameters for deep local exploitation, and a multiple-insertion (MI) method used for re-initialization in the scout bee phase. Many instances from a real factory were used as test benchmarks. Comprehensive analysis of the experimental results showed that the proposed DABC algorithm was significantly superior to the four metaheuristics in the literature.