| Summary: | 碩士 === 國立成功大學 === 工程管理碩士在職專班 === 105 === Die-cutting is a common process of reel processing in the manufacturing industry, and it can continuously generate products with pre-determined shapes. Currently, to the best of our knowledge, there are no methods to optimize the design of die-cuts for angle constrained optical film products; in particular, the units in the die-cuts must be arranged in a specific angle.
In this study, a mixed integer programming (MIP) and a particle swarm optimization (PSO) algorithm are proposed by using mathematical programming technique and meta-heuristic algorithm respectively. In order to verify the models, an empirical study of Taiwan’s display manufacturers is conducted to solve the die-cuts with 6 rectangles and 12 rectangles, and compare with the results of the existing artificial arrangements. It is found that the mixed integer programming model can produce better results in the two die-cuts examples, which can increase the utilization rate of 3.41% and 1.2%, respectively; and the particle swarm algorithm model can produce a result which is similar to the result of the artificial arrangement in the die-cut with 6 rectangles. But there is no better solution to the die-cut with 12 rectangles in a limited time spent.
The results show that the models constructed in this study are feasible to design die-cuts, and can be used for practical implement in the near future. It is expected to reduce waste and save costs, and thus enhance the industrial core competence.
|
|---|
