本研究主要在探討當存貨品質具有退化的特性時，如何設定適當的緩衝，同時結合派工法則與檢驗計劃，以確保瓶頸作業的順暢並進而提升產出。針對瓶頸檢驗站前的緩衝區管理，本研究發展出派工法則來控制投料以確保緩衝區的固定大小與內容，同時改善瓶頸的保護與防止在製品因退化所形成的回流。研究的對象為某晶圓廠在潔淨室（Clean Room）中的最終檢驗站CR-80，該檢驗站為瓶頸工作站，由於在原來的作業設定下，部分產品檢驗採用抽檢，抽檢不通過即進一步實施全檢，使得作業時間存在著相當的變異。而經過清洗後置於緩衝區內等待檢驗的晶圓，隨著時間所附著的污染微粒（Particle）逐漸增加，品質發生退化的現象，退化至一定程度時，晶圓就必須重洗。因此，加工時間的變異以及回流重工的問題，影響了系統的產出。研究結果顯示，在維持原先的抽檢計劃下，運用本研究所發展出的作業方法對於系統產出率有9%-15%的改善，而瓶頸設備CR-80的使用率亦提升5﹪-15%。

The aim of the research, under deteriorating inventory condition, is to develop proper shop floor management guidelines that coordinate various order release mechanisms, sampling plan schemes and dispatch rules to smooth the bottleneck operation and to improve the system performance. The subject of this research is the final inspection station CR-80, which is the bottleneck and located in a class 10 clean room. The washed wafers are put into the buffer and wait for inspection, particles adhere to the wafers as queue time goes by. The quality of wafers deteriorates and causes unavoidable reflow of WIP. Currently, a sampling inspection plan is used to ensure the outgoing quality of of wafers. Once a lot failed to pass the sampling inspection, a full inspection is conducted immediately. Such inspection scheme varies the processing time greatly and introduces random behavior, which in term affects the system throughput. A dispatching rule based on minimization of process time variation was developed and several simulation experiments were conducted. The comparisons between the simulation results with the real life data indicated that without changing current sampling plan, the new dispatching rule improved about 10 % of the throughput with little increase in number of reflow. The degree of improvement of machine utilization varied under different sampling plans, within the range of 5 to 15%.

目錄
致謝 I
摘要 II
Abstract III
目錄 IV
表錄 VIII
圖錄 IX
符號說明 X
第一章 緒論 1
1.1研究背景 1
1.2 研究動機 2
1.3 研究目的 3
1.4研究範圍與限制 6
1.4.1 研究範圍 6
1.4.2 研究限制 6
1.5研究流程 7
1.6論文架構 8
第二章 相關文獻探討 9
2.1在製品（WIP）數量的控管 9
2.2限制理論（TOC） 10
2.3投料與派工策略 12
2.3.1均勻負荷法（Uniform Loading ; UL） 13
2.3.2避免飢餓法（Starvation Avoidance ; SA） 14
2.3.3雙界法（Two-Boundary ; TB） 15
2.3.4工作負荷調節法（Workload Regulating ; WR） 15
2.3.5及時生產法（JIT1）與修正即時生產法（JIT2） 16
2.3.6定量在製品法（Constant WIP ; CONWIP） 16
2.3.7鼓-緩衝區-繩投料法（Drum-Buffer-Rope） 17
2.4抽樣檢驗 17
2.4.1抽樣檢驗的基本名詞 20
2.4.2抽樣檢驗計劃的形式 21
2.4.3操作特性曲線（Operating Characteristic Curve） 25
2.4.4選別型抽樣檢驗計劃 27
2.5退化性存貨相關研究探討 28
第三章 研究方法 31
3.1情境假設 31
3.1.1晶圓等候情境假設 31
3.1.2最終檢驗機台判斷方式情境假設 32
3.2相關的作業指導原則發展 33
3.2.1投料方式與緩衝區大小控制 33
3.2.2檢驗站抽樣檢驗作業方式 34
3.2.3 派工策略 35
第四章 模型建構 36
4.1模型建構步驟 36
4.1.1基本模型問題界定與假設 37
4.1.2模型一般狀況設定 38
4.1.3模型1基本假設 40
4.1.4模型2基本假設 42
4.1.5基本模型模擬結果分析 44
4.2抽樣與派工模型 45
4.2.1抽樣模型的問題界定與假設 45
4.2.2派工模型的問題界定與假設 48
4.2.3抽樣及派工模型建構 49
4.2.4模型A1與模型A2基本假設 51
4.2.5模型A3與模型A4基本假設 52
4.2.6模型A5與模型A6基本假設 53
4.2.7抽樣及派工模型模擬結果分析 54
4.3總面積良率遞減模式模型假設 55
4.3.1總面積良率遞減模式模型建構 56
4.3.2總面積良率遞減模型模擬結果分析 56
4.4條件的寬放 58
4.4.1寬放模型的建構 58
4.4.2模型C1及模型C2基本假設 61
4.4.3模型C3及模型C4基本假設 62
4.4.4模型C5及模型C6基本假設 63
4.4.5模型C7及模型C8基本假設 65
4.4.6模型C9及模型C10基本假設 66
4.4.7模型C11及模型C12基本假設 67
4.4.8寬放模型模擬結果分析 68
第五章 結論 70
5.1結果與討論 70
5.1.1 TOC環境下退化性存貨作業指導原則 72
5.2未來研究方向建議 73
參考文獻 74
附錄一 基本模型程式碼 79
簡歷 94

中文部分
[1] 大野耐一，1981，豐田生產方式與現場管理，日本能率協會，中華企業管理發展中心。
[2] 呂金隆，2001，晶圓製造廠瓶頸管理模式之研究，華梵大學工業管理研究所碩士論文。
[3] 呂柏賢，2000，灰色需求預測模式之研究-以易腐性商品為例，東海大學工業工程研究所碩士論文。
[4] 李慶恩、林於杏，1998，「考慮加工時間變異之晶圓製造在製品訂定模式」，工業工程學刊，頁255-267。
[5] 李榮貴，1992，「限制理論－製造管理的新觀念」，機械工業雜誌，頁212-223。
[6] 林明獻，2000，矽晶圓半導體材料技術，全華科技圖書股份有限公司。
[7] 張有成，1999，抽樣檢驗，中華民國品質學會，第六版。
[8] 張盛鴻、張立賢、李榮貴，2001，「晶圓製造廠之整合式生產控制方法」，工業工程學刊，頁59-72。
[9] 黃惠民、鄧貴中，1995，「韋伯分配型損耗性產品在儲存場地及存貨投資金額限制條件下之最佳生產及定價策略」，中原學報。
[10] 鄭春生，1998，品質管理，育友圖書有限公司。
[11] 鍾淑馨、黃宏文，1998，「晶圓製造廠生產作業控制策略之設計」，工業工程學刊，頁93-113。
英文部分
[12] Aggarwal, A., J. Park, 1993, “Improved algorithm for economic lot size problems,” Operations Research, Vol.41, pp.549-571.
[13] Brant, E. F., 1975, “Optimal Ordering Policy for a Perishable Commodity with Fixed Lifetime,” Operations Research, Vol.23, No.1, pp.46-61,.
[14] Cohen, M.A., 1977,“Joint Pricing and Ordering Policy for Exponentially Decaying Inventory with Known Demand,” Navel Research Log. Quag Vol.24, No. 1, pp. 257-268.
[15] Cover, R.P. and Philip G.C., 1973, “An EOQ Model for Item with Weibull Distribution Deterioration,” AIIE Transactions, Vol.5, No.4.
[16] Federgruen, A., M. Tzur. 1991, “A simple forward algorithm to solve general dynamic lot sizing models with n periods in O(n log n) or O(n) time, ” Management Sci. Vol.37, pp.909-925.
[17] Friedman, Y., Y. Hoch. 1978, “A dynamic lot-size model with inventory deterioration, ” INFOR Vol.16 pp.183-188.
[18] Glassey, C. R., and Resende, M. G. C. ,1998, “Closed-Loop Job Release Control for VLSI Circuit Manufacturing,” IEEE Transactions on Semiconductor Manufacturing, Vol. 1 No. 1, pp.36-46.
[19] Goldratt E.M. , 1990, The Haystack Syndrome, North River Press Inc.
[20] Jain, K. and A. S. Edward, 1994, “Lot sizing for a Product Subject to Obsolescence or Perishability” European Journal of Operation Research, Vol.75, pp. 287-295.
[21] Lawton J. W. et. al., “Workload Regulating Wafer Release In a GaAa Fab Facility,” Int’l Semiconductor Manufacturing Science Symposium, pp. 33-38.
[22] Levitt M. E., and J.A.Abraham, 1990, “Just-In-Time Methods for Semiconductor Manufacturing,” IEEE/SEMI Advanced Semiconductor Manufacturing Conference, pp.3-9.
[23] Lou S., and P.W. Kager, 1989, “A Robust Production Control Policy for VLSI Wafer Fabrication,” IEEE Trans. On Semi. Manu., Vol. 2, no. 4, pp. 159-164.
[24] Misra, R.B., 1975, “Optimum Production Lot Size Model for a System with Deterioration Inventory,” Int. J. Prod. Res., Vol. 13, No. 5, pp. 495-505.
[25] Nahmias, Steven, 1976, “Myopic Approximations for the Perishable Inventory Problem, ” Management Science, Vol.22, No.9.
[26] Nahmias, S. 1982, “Perishable inventory theory: A review, ”Oper. Res. Vol. 30 pp.680-708.
[27] Park, S.P, 1983, “An Integrated Production-Inventory Model for Decaying Raw Materials,” Int. J. System SCI., Vol. 4, No. 7, pp. 801-806.
[28] Philip, G.C, 1974, “A Generalized EOQ Model for Item with Weibull Distribution Deterioration,” AIIE Transaction, Vol. 5, No. 4.
[29] R.Michael Mahoney, 1997, “High-Mix Low-Volume Manufacturing,” Hewlett-Packard Professional books.
[30] Spearman M.L., and D.L. Woodruff, 1990, “CONWIP : a pull alternative to kanban,” Int. J. Prod. Res., Vol. 28, no. 5, pp. 879-894.
[31] Wallace J. Hopp, and Mark L. Spearman, 2000, “Factory Physics,” Irwin McGraw-Hill.
[32] Wagelmans. A. P. M., S. Van Hoesel, A. Kolen, 1992, “Economic lot sizing: An O(n log n) algorithm that runs in linear time in the Wagner-Whitin case, ” Oper. Res. Vol.40 pp.145-156.
[33] Wee, H.M, 1993, “ Economic Production Lot Size Model for Deteriorating Item with Partial Back-Ordering,” Computers & Ind. Engng, Vol. 24, No. 3, pp.449-458.
[34] Wee, H.M, 1993, “Perishable Commodities Inventory with Patial Backordering,” Chung Yuan Journal, Vol. 24, No. 3, pp. 449-458.
[35] Xu, Hai-Ping and Wang, Hsu-Pin, 1992, “Optimal Inventory Policy for Perishable Items with Time Proportional Demand, ” IIE Transaction, Vol. 24, No.5, pp.105-110.
[36] Yan H., S. Lou S. S. A. Gardel, and P. Deosthali, 1992, “Testing the Robustness of Various Production Vontrol Policies in Semiconductor Manufacturing,” IEEE Trans. on Semi. Manuf., pp. 1-24.
[37] Zabel, E, 1964, “Some generalizations of an inventory planning horizon theorem,” Management Sci. Vol.10 pp.465-471.
[38] Zangwill, W. I. 1966, “A deterministic multi-period production scheduling model with backlogging,” Management Sci. Vol.13 pp.105-119.
[39] Zangwill, W. I.1969, “A backlogging model and a multiechelon model of a dynamic economic lot size production system─a network approach,” Management Sci. Vol.15 pp.506-527.