以蟻群演算法探討模具製造之排程最佳化

塑膠早已廣泛應用於各種產品，尤其3C產品快速的發展，模具的開發速度便愈來愈受到重視；而模具製造排程屬於零工式排程問題，許多工廠早已採用工作站的方式，來增加整體產能。如果還使用傳統的單機排程方式，已經無法有效地利用生產資源，故應從平行機台的角度，妥善規劃排程與機台組態，減少模具製造時間與資源用量，才是增加企業競爭力的關鍵。在工作站內會有許多平行機台，同樣工作可以有多部機台供選用，此狀況下如何選擇適當機台成為一大難題，本研究利用蟻群演算法的求解能力，建構以蟻群演算法為基礎的模具製造之平行機台排程系統，針對平行機台的特性，設計雙層的節點架構，讓螞蟻選擇製程節點後，再選擇機台節點，藉由費洛蒙的累積與蟻群系統三大法則的運作，能有效的安排工作至適當的機台。針對電極塊與放電加工的部份，加入節點篩選機制，設計多對一之節點架構，確保放電加工在電極塊製程之後。另外在機台配置方面，雖然增加機台數量能使排程調度更具有彈性，卻也增加了企業的成本，為了達到理想的機台組態，本研究搭配田口方法，將機台數量設定為控制因子，利用直交表來導出各因子的品質特性，其中影響較大的因子即為瓶頸機台，改變其水準值來得到較理想的完工時間，並進一步討論各設備的數量對製造排程的影響，評估工廠生產績效對於生產資源數量之關係。在製造排程引導流程系統的架構下，透過介面的操作來完成各參數的設定，並根據演算法的結果提供機台組態的建議值，提升機台的產能，最後的排程結果可輸出成甘特圖，並讓生管人員做最後的檢查與調度，使工廠的流程更為順暢。

關鍵字

零工式排程問題；最佳化；蟻群系統演算法；模具製造

並列摘要

Plastics are used in an enormous and expanding range of products, especially consumer electronics, mold development speed will be more and more important. Mold manufacturing schedules are more complex Job-shop Scheduling Problem. Many factories already adopted work stations to increase the overall production capacity. Still use conventional job-shop scheduling system that can be applied well to multi-stage, single machine producing mode is not an effective way to use production resources. Consider parallel-machine producing mode and make appropriate scheduling, effectively shorten the manufacturing time can enhance market competitiveness. Same work can choose multiple machines, how to select appropriate machine under this condition is a big challenge. This research utilizes ant colony system with their exceptional searching and solving function to construct a modeling and scheduling system for mold manufacturing, parallel-machine. Design double node architecture, allowing ants to select process node, then choose the machine node, by the cumulative pheromone and three rules in ant colony system, effective arrangements jobs to the appropriate machine. As to electric discharge machining and electrode block, join node filter mechanism, design many-for-one node sequence architecture, ensure that the discharge machining process after electrode block process. In the machine configuration, increase the number of machines can make scheduling more flexible, but also increases the resource costs. In order to achieve the desired machine configuration, Taguchi method used in this study. The number of machine is set to control the factors, using orthogonal arrays to export quality characteristics of each factor, factors which affect the larger is the bottleneck machine, change its level to get better makepan. Further discuss the impact of the number of each machine manufacturing scheduling, assess the performance of the plant for the production relations of production quantity of resources. Under the guiding procedure for manufacturing systems, through the user interface to set the parameters, and provide recommendations based on the results of algorithm to enhance energy production machine. The results can be output to a scheduling gantt chart, allowing managers to do a final check and scheduling, the plant processes more smoothly.

並列關鍵字

Planning and Scheduling ； Mold-Manufacturing ； Optimization ； Ant Colony System

參考文獻

[2] Dorigo, M. & Gambardella, L. M. (1997). Ant Colony System: a Cooperative Learning Approach to the Traveling Salesman Problem. IEEE Transactions on Evolutionary Computation, 1(1), 53-66.

[4] Palmer, D.S. (1965). Sequencing Jobs Through a Multi-stage Procrss in the Minimun Total Time - A Quick Method of Obtaining a Near Optimum. Operational Research Quarterly, 16(1), 101-107.

[5] Gupta, J.N.D. (1971). A Functional Heuristic Algorithm for the Flow-shop Scheduling Problem. Operational Research Quarterly, 22(1), 39-47.

[6] Chou, Y. C. & Wu, C. S. (2002). Economic Analysis and Optimization of Tool Portfolio in Semiconductor Manufacturing. IEEE Transactions on Semiconductor Manufacturing, 15(4), 447-453.

[7] Chang, P. T., Lin, K. P., Pai, P. F., Zhong, C. Z., Lin, C. H. & Hung, L. T. (2008). Ant Colony Optimization System for a Multi-Quantitative and Qualitative Objective Job-Shop Parallel-Machine Scheduling Problem. International Journal of Production Research, 46(20), 5719-5759.

國際替代計量

以蟻群演算法探討模具製造之排程最佳化

全文下載

主題瀏覽