A hybrid approach using ant colony optimisation for integrated scheduling of production and transportation tasks within flexible manufacturing systems
研究了柔性制造系统中生产与运输任务的集成调度问题，提出一种基于蚁群优化的混合方法，通过两元素向量结构建模决策节点，并融入启发式规则避免局部最优，数值测试表明该方法优于现有元启发式方法。
This paper studies the integrated scheduling problem in flexible manufacturing systems (FMS), where flexible machines and Automated Guided Vehicles (AGV) shared by production jobs are scheduled simultaneously in an integrated manner. Routing flexibility, a crucial advantage of FMS, enabling a job to be handled via alternative machine combinations, is involved. To address this problem, we propose a novel hybrid approach using Ant Colony Optimisation (ACO), which employs a two-element vector structure to model the ACO decision nodes. Each node represents an operation from a job assigned to a particular machine. During the ACO process, to decide a node for next movement, an ant first assesses potential nodes through a node scheduling procedure with two consecutive steps: firstly, using a heuristic vehicle assignment method, an AGV is designated and scheduled for the operation specified in a node. Following this, based on the established transportation timeline, the operation’s production schedule on the assigned machine is determined. Subsequently, the node selection is guided by the pheromone information on potential paths and the heuristic data of potential nodes derived from their scheduling information. To avoid local optima, multiple heuristic rules are incorporated in the ACO, with one chosen randomly for node selection each time. Numerical tests show that our proposed approach outperforms contemporary metaheuristic approaches in the literature. In addition, its efficiency of handling complex problem instances is also assessed and demonstrated.