Research On Operation And Control Of Shop-floor Flow Manufacturing For Application In High-mix And Low-volume Production

Nowadays, variable demands of customers have gradually made the production mode shift from high-volume and repetitive to high-mix and low-volume. The uncertainty of demands and turbulence in production process have brought great challenges to the operation management in high-mix and low-volume environment. Flow manufacturing technology enables manufacturing various types of similar products with different volume mix in the same production system (or cell), with seldom or just a few changes to the current production environment. However, the traditional flow manufacturing technology is not adaptable to the production environment with frequently changed production mix and higher demand variability. Therefore, a new mode of flow manufacturing as well as its operation and control method in high-mix and low-volume environment are discussed in this paper.Firstly, this thesis provides a systematic literature review on the methods for operation and control in manufacturing system and its related fields, and points out the characteristics and difficulties in the high-mix and low-volume production environment. The differences of flow manufacturing between the traditional environment and high-mix environment are analyzed. Based on this, the operation framework for high-mix and low-volume production is established.Then, the related techniques in the different levels of the operation framework are studied in detail as follows:(1) In the overall operation and control level, the dynamic formation and management based on virtual manufacturing cell are presented. It simplifies the material flow and makes the production system easy to manage. Meanwhile, in combination of the machine sharing, it improves the utilization of equipments, which is suitable to the variable demand in high-mix and low-volume production. The flow path design process is divided into two stages. Part families are formed in the first stage. A mathematical model based on similarity coefficient is established. In the second stage, machines are allocated to the flow paths corresponding to part families. The machine sharing is feasible by using a novel mathematical programming approach. The objective is to maximize the throughput of the system and minimize the number of shared machines. It addresses the machine sharing, and setup times on each machine are also considered.(2) To balance the workload on a flow paths with machine sharing based on finite capacity, the reasonable workload allocation on the proper resource in the proper time bucket is studied. Considering flexible routing and setup times, a mathematical model is established to describe the workload allocation problem. This model describes different group rules on different machines and estimates the setup times reasonably. The alternate resources are considered when corresponding operation to resources. All these factors make this model practical, especially to the flexible job shop in high-mix and low-volume production. Besides, a heuristic algorithm is proposed to deal with the large-scale problem. This algorithm initiates with the relaxation of capacity constraint, and then adjusts the workload by offloading, pulling and pushing, to make capacity constraint satisfied again. This algorithm reaches a good solution at an acceptable speed for the real shop floor.(3) As to the operation and control of cellular level in a specific flow path, the production scheduling and control methods based on theory of constraints (TOC) are studied. The thesis emphasizes on the group scheduling problem on bottleneck. Group scheduling decreases the setup times so as to improve the utilization of bottleneck. A heuristic algorithm with the objective of minimizing the maximum tardy time is utilized to balance the conflict between the setup time and due date. Then, three different material release control policies are analyzed and buffer management is discussed.After that, to validate and demonstrate the application of the results above, the proposed new mode of operation and control has been preliminarily implemented in a tube manufacturing shop of a shipyard based on its production conditions and requirements. It tries to improve the shop floor by using an advanced planning and scheduling system X-Planner which is developed independently. The expected application effect of the improvement is verified by a simulation system.Finally, some conclusions are drawn with some of the further research directions being anticipated.