| Summary: | 碩士 === 國立雲林科技大學 === 工業工程與管理系 === 104 === To satisfy the needs to supply and develop potential possibilities, industries urged to expand their production capacity. While the original plants can’t wait to be expanded, industries would thrive to meet the production capacity via setting up new plants or merge other small scales, then become a multi-plant production pattern.
To see through the difference between multi-plant scheduling and traditional one, the production ability, limitation of production capacity, and the way it differs from the cost to produce the same or totally distinct products in different plants are need to require more attention; in this case, the scheduling could be more complicated. This research was done based on the problem to the needs of containing relatively multi-product in multi-plant, in each plant, it has own production conditions and costs. Considering every order demand and due window to model the mathematical model of the multi-production assigning and scheduling in an environment of make to order. Furthermore, this research is to thought out the best assignment of upcoming order minimizing the sum of production cost, early cost and tardiness cost; the real case is a machine tool parts manufacturer. For current mode with not split mode and forced split mode to do performance evaluation. By Particle Swarm algorithm for assigning the results of this study showed that when more orders or when capacity is insufficient, assigned to order segmentation model can have total costs of smaller, conversely, don't split mode can have better performance, enterprises can depend on the actual order or production conditions, choose whether you want to split the order.
|
|---|
