請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/32906
標題: | 12吋晶圓製造廠自動搬運系統之
最佳化派工與排程 Optimized Dispatching and Scheduling for OHTs in a 300mm Wafer Fab |
作者: | Da-Wei Chan 詹大瑋 |
指導教授: | 傅立成(Li-Chen Fu) |
關鍵字: | 十二吋晶圓廠,排程,派工,裴氏網路, 300mm wafer fab,scheduling,dispatching,Petri net, |
出版年 : | 2006 |
學位: | 碩士 |
摘要: | 在這篇論文裡,我們利用一圖形及數學之具有時間和顏色屬性的裴氏網路來建構晶圓廠之生產流程。而這篇我們以懸吊式搬運車輛作為搬運工具,並且利用優先權法則來達成機台派工與車輛派工。此外,我們應用基因遺傳演算法來幫助我們獲得一進似最佳解。
在這篇論文中的模型由兩部份組成,包括生產模型以及運輸模型。利用這個具有時間和顏色屬性的裴氏網路,我們可以模擬生產步驟和運輸的問題,因此機台狀態,批貨狀態都可以有效率且正確地處理。此外,不同的排程方法可以經由模擬來評估,則其中較好的的方法即可以找到。在批貨派工的派送過程中,我們首先將其分為以批貨為考量的機台選擇和以機台為考量的批貨選擇。空閒的機台會從多個批貨中找出最合適的批貨來處理,而批貨會找出最合適的機台加工。在懸吊式搬運車輛派送上,我們會依目前交通的狀況來重新對懸吊式搬運車輛與批貨作配對。 此外,我們利用”區段控制”來確保懸吊式搬運車輛不會發生碰撞,以及”推進車輛”來防止阻礙的情形。此外,利用車輛重新分配可以減少阻礙的情形,且交通擁塞的情形也可以被考量並減低。 在最後排程階段,我們利用基因遺傳演算法來獲得近似最佳解。從我們的實驗結果可以看出,由時間和顏色屬性的裴氏網路為基礎的基因遺傳演算法可以產生出有效率的解,故這個基因遺傳演算法排程確實可以適應環境的快速變化,像是半導體晶圓製造廠。 In this thesis, we use a graphical and mathematical modeling tool- Coloured Timed Petri Net (CTPN) to model the production flow in the wafer fabrication plant. Overhead hoist Transport (OHT) is taken as the transportation vehicle, and we take the priority rules for lot dispatching and OHT dispatching. Moreover, we apply the genetic algorithm to help us obtain the near-optimal solution. The model in this thesis comprises two parts, including the processing model and transportation model. With this CTPN model, we can simulate the production process and transportation issues, thus the equipment status, OHT status, and the lot conditions can be tracked efficiently and precisely. In addition, different scheduling policies can be evaluated via simulations and a superior policy will then be determined. In the dispatching phase of lot dispatching, we first present the lot based selection scheme and the equipment based selection scheme. The available equipment will select the fittest lot according to the lots priority if there are multiple waiting lots, and the lot will choose the fittest equipment under multiple pieces of available equipment. In the dispatching phase of OHT dispatching, reassignment of the pair of lot and OHT is applied as the traffic condition of the plant is changing. Besides, “zone control” is used to ensure that OHTs will not collide, and “push vehicle” is used whenever an idle OHT blocks the zone another OHT is trying to occupy. Moreover, through OHT reassignment the block phenomenon can be eliminated, and the traffic congestion problem is also considered and reduced. At last in the scheduling phase, we apply the genetic algorithm (GA) based approach to obtain a near-optimal solution to our scheduling problem. From our experiment results, the developed CTPN based genetic algorithm will yield a more efficient solution than other scheduler, so that this GA scheduler can indeed satisfy the need of a rapidly changing environment, such as the wafer fabrication plant. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/32906 |
全文授權: | 有償授權 |
顯示於系所單位: | 資訊工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-95-1.pdf 目前未授權公開取用 | 689.18 kB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。