Current manufacturing systems have a tendency to become more and more flexible to adapt to the needs of product diversification. Such a system, e.g. flexible manufacturing system, consists of a number of automatic machines, material handling devices such as automated guided vehicles (AGVs) or mobile robots, and a central control computer. Mobile robots, as well as AGVs, can move around in their working space to transport components among machines. The main novelty of this work is that mobile robots can also execute various value-added tasks without human intervention thanks to their manipulation arms which is not possible for AGVs. This new characteristic certainly makes the problem more complex and computationally expensive. To utilize these manufacturing systems in an efficient manner, it is necessary to schedule transportation of product components by mobile robots and to schedule processing of products on machines possibly by mobile robots. Consequently, the key innovation of this work lies in consideration of three interrelated sub-problems which must be solved. They include computing the sequence of operations on machines, the robot assignment for transportation, and the robot assignment for processing. To achieve this goal, a computationally efficient hybrid heuristic method combining genetic algorithm and tabu search is developed to solve the problem considering makespan minimization. A mixed-integer programming (MIP) model is formulated. Another combined method using results of the hybrid heuristic is proposed to speed up the solving of MIP model. The quality of hybrid heuristic’s solutions is compared and evaluated by using those of the MIP model as reference points.

中文翻译：

---

安排用于运输和制造任务的移动机器人

当前的制造系统趋向于变得越来越灵活以适应产品多样化的需求。这样的系统，例如柔性制造系统，由许多自动机器，诸如自动导引车（AGV）或移动机器人之类的材料处理设备以及中央控制计算机组成。移动机器人以及AGV可以在工作空间中移动，以在机器之间运输组件。这项工作的主要新颖之处在于，由于其操纵臂，移动机器人还可以执行各种增值任务而无需人工干预，这对于AGV来说是不可能的。这一新特性无疑使问题变得更加复杂且计算量大。为了有效利用这些制造系统，必须通过移动机器人安排产品组件的运输，并可能通过移动机器人安排机器上产品的处理。因此，这项工作的关键创新在于必须解决三个相互关联的子问题。它们包括计算机器上的操作顺序，用于运输的机器人分配以及用于处理的机器人分配。为了实现这一目标，开发了一种将遗传算法和禁忌搜索相结合的计算有效的混合启发式方法，以解决考虑到制造时间最小化的问题。建立了混合整数编程（MIP）模型。提出了使用混合启发式结果的另一种组合方法，以加快MIP模型的求解速度。