透過您的圖書館登入
IP:3.145.105.108
  • 學位論文

考量等候時間限制之多階製造鏈整合

Integration of Multi-stage Manufacturing Shops with Queue-time Constraint

指導教授 : 周雍強

摘要


半導體製造系統複雜,尤其在今12吋晶圓廠的時代,整個生產階段形成專業分工的情形,各生產節點各自做好最好的生產規劃,但是在跨廠區之間卻缺乏一個良好的系統來整合,因此有許多的問題需要解決,其中之一便是等候時間限制。在特定製程間會有等候時間之限制,前製程完成後需於特定等候時間限制內進入後製程,超過此一時限會產生品質問題,需進行重工損耗產能。 有許多的文獻在處理多零工式生產整合的議題上,是使用數學規劃法或是模擬法來求取最佳解,但是當問題的複雜度變大時,決策時間會呈倍數成長,不適合短期、動態的環境,有鑑於此,本研究發展一個以靜態產能估計為基底的整合方法(Queue estimate-based Lot Permission rule; QEBP),係利用目前可得之資訊,如工件規格、現場狀況等資訊去做組合運算,以快速的估計生產週期時間。 此資訊可用來處理工單有等後時間限制的特性;上游的生產節點可以經由預估下游暫存區的在製品負荷長度,然後在適當的時間點投料,避免超出等候時間限制的情形出現,經過模擬實驗的結果分析,發現QEBP在不同的環境設置下,均比完全放行法則,更能提高產量、有效率的估計產能負荷,能幫助管理者快速的做出決策。

並列摘要


The semiconductor manufacturing system is complex, each of manufacturing node do the best production planning, but lack of cross cooperation between the factory. There are many factors affecting the throughput and yield rate. One of them is the “Queue-Time-Limit”. Certain process steps have the queue-time-limit, which means a lot needs to begin the following process step within certain time limit after it completes its operation of the previous process step. If a lot exceeds the queue-time-limit, quality issues will occur and the lot needs to be reworked Many of the references deal with multi job shops integration issues, they use mathematical programming method or the simulation method to obtain the optimal solution, but when the complexity of the problem become larger, decision time will grow exponentially. This situation is not suitable for short-term,dynamic environment. So, this research develop a static capacity estimated based integration method (Queue estimate-based Lot Permission rule; QEBP) it using the available information currently, such as the lot type, site of conditions or so on to Combination calculation, in order to estimate the production cycle time quickly. This information can be used to handle work orders with the queue time limit characteristics upstream production node estimate the downstream buffer loading, and then feed the lots at appropriate time to avoid waiting time exceed the limit .After the simulation analysis, we found QEBP complete more work orders than all-release method in different environment ,it’s also estimate the capacity loading efficiently that can help managers make decisions quickly.

參考文獻


[4] 葉家佑(2005),產品組合對平行機台工作分派效率之影響, 台灣大學工業工程研究所碩士論文
[6] 黃嘉常(2007), 考量等候時間限制之半導體製造在製品分配及控制方法 , 清華大學工業工程研究所碩士論文
[5] 巫啟彰 (2006),半導體晶圓製造廠中等候線間限制作業的放行控制方法, 清華大學工業工程研究所碩士論文
[7] Berning, G., M. Brandenburg, et al. (2004). "Integrating collaborative planning and supply chain optimization for the chemical process industry (I)--methodology." Computers & Chemical Engineering 28(6-7): 913-927.
[8] Chen, W.-L., C.-Y. Huang, et al. (2009). "Multi-tier and multi-site collaborative production: Illustrated by a case example of TFT-LCD manufacturing." Computers & Industrial Engineering 57(1): 61-72.

延伸閱讀