臺灣博碩士論文加值系統

由於市場競爭日益激烈，消費者意識覺醒與對服務滿意度的要求增加，使得一般傳統的販賣銷售、低價競爭行為，也漸漸無法滿足消費者的需求。同時，這樣沒有差異化的服務提供也無法成為企業的競爭優勢。當然，在提供服務的同時，為了避免消費者無法在第一時間獲得即時的服務與產品。企業通常會透過存貨的儲備，來達到滿足顧客的及時且不確定的需求。因此，也造成企業對於商品或是其他產品服務該如何衡量存貨水準的大小產生疑惑，本研究即以存貨管理作為出發，並在考慮需求的變異已知下，針對供給上前置時間的不確定性進行探討。
國內的汽車產業，一向不具有生產上的規模經濟，生產批量小相對造成對上游供應商的議價能力不足。除了必須支付其母廠所授權代理的權利金之外，市場佔有率也受到國內外不同車種的蠶食，加上大量非價格競爭的廣告費用支出，更侵蝕了原本已漸趨微薄的獲利能力。
因此，各車廠除了利用價格與贈品優惠鼓勵買氣，也轉而進入汽車銷售後市場，提供維修服務、車輛保險經營以及銷售分期租賃服務予顧客。其中，提供現有車主完善的汽車維修服務，使得顧客能即時即地，享受高品質的維修服務，更能確保顧客未來的回流率。同時，希望透過一定水準的零件備料，能夠滿足顧客，維持服務水準。但是，過多的存貨積壓企業的流動資金，不足的存貨又損害服務水準。本研究目的即在於如何透過企業現有的存貨管理基礎上，考慮供應端在不確定性的前置時間供應下，如何估計個別零件因為隨機性的前置時間所造成的存貨。本研究欲利用FCM分群演算法，透過零件的供應特性建立分群原則，並且使用相關影響存貨的成本權衡關係，利用分群模式，討論各群聚下最適之訂購週期、前置時間決定以及因為供應不確定下所必須額外準備的安全存貨。
研究中利用eM-Plant針對實際資料進行模擬驗證，並挑選分群結果中變異係數最大與最小的兩群作實證分析，由結果證明本研究所進行之存貨政策，並且符合在個案公司現行的服務水準限制下，本研究建置之模型可以有效改善現行個案公司所運作之存貨水準計算。而且增加成本面的經濟效益，以避免大量存貨下耗費之存貨的機會成本，以及存貨過少時所額外花費的緊急運輸成本與短缺成本。並且由敏感度分析得到，若零件在啟動緊急空運下，非為整裝貨櫃運送，則運輸成本增加的幅度對訂購週期之影響。另外，本研究同樣對納入其他費用於存貨持有成本計算下，對訂購週期之增減效果影響。

Due to the globalization and the World Trade Organization (WTO) participating regulation, Taiwan has lowered automobile tariff year by year and also canceled the limitation of spare parts homemade rate of homebred autos. Viewing the local market, local automotive companies have suffered in dilemma of price war to protect their market shares and faced the challenge from importing cars. Without economic of scale in production, local agents have no bargaining power to suppliers; moreover, large amount of adverting and authorized expenses also erode the profitability of local agents. All of these lead automotive industry in Taiwan to a tough situation.
In spite of diminishing of car selling market in Taiwan since 1994, enterprises need to develop another department to recover the lost margin in car selling. Local agents turn into after-sales service market to provide diversified service, such as repairing service, insurance, purchasing installment and auto releasing service, not just offering a premium and additional dividends to promote car selling. In order to ensure customers satisfaction to re-consume, local agents must store a fixed level of spare parts to meet customers’ emergency demand and maintain a certain service level. Too much and unnecessary inventory may keep too much current capital, resulting in difficulty of business operating. Lowering down inventory level , however, may damage the desired service level. In this paper, it is studied how to restructure a new inventory policy based on the existing inventory management methods although there are plenty of papers about the relationship of inventory level arising and stochastic demand.
This study focus on discussing the stochastic of lead time of supply-side generate rising inventory under known and controllable demand variation. By taking a real local automotive agent—H Inc. as the case to discuss, this study plans to use FCM algorithm for clustering the parts most frequently demanded with large quantity. Through setting up clustering rules, the optimal reorder cycle would be calculated by the tradeoff among related inventory costs. And at last, it turns out to establish a reasonable and logical inventory policy for H Inc. used.
As the results of simulations, it is found that programming model of this study can efficiently decrease current inventory quantity under certain limitation of service level. The difference between air and sea transportation expenses and the margin of service parts has mutually impact on deciding order cycle. Further via sensitive analysis, it is suggested for enterprise to lift order cycle if the less-than-cargo emergency parts transportation expenses increase or inventory carrying costs raises. Observing from the inventory model, moreover, other inventory-related opportunity cost would have great influence on inventory level for the purpose of maximizing customer satisfaction.

致榭 I
摘要 11
Abstract IV
目锋 IV
表目锋 IX
圆目锋 X
第一章 緒論 1
1.1 研究動機 1
1.2 研究目的 3
1.3 研究架構 4
1.4 論文架構 6

第二章 文獻探討 8
2.1 存貨模型相關文獻探討 8
2.1.1 供應鏈與物流管理 8
2.1.2 存貨之定義與功能 8
2.1.3 存貨分類與存貨管理 10
2.1.3.1 存貨種類 10
2.1.3.2 存貨管理 10
2.1.4 存貨管制模式 12
2.1.4.1 永續盤存制 13
2.1.4.2 定期盤存制 15
2.1.4.3 其他存貨管制模型 16
2.2 分群演算法相關文獻探討 17
2.2.1 分群之定義與功能 17
2.2.2 分群演算法 17
2.3 供應端之隨機性前置時間相關文獻探討 20

第三章 模型設計 22
3.1 汽車零件市場現況分析 22
3.2 模型概念與分析架構 25
3.3 模型基本假設 27
3.4 分群模式與訂購參數決定 28
3.4.1 分群符號說明 29
3.4.2 分群模型參數計算說明 30
3.4.3 分群演算法及其應用 32
3.5 成本最適化模型 37
3.5.1 成本模型符號說明 37
3.5.2 成本模型參數計算說明 39
3.5.3 成本模型之建立 41
3.5.3.1 成本最適模型 41
3.6 執行步驟 46

第四章 個案研究與實證分析 49
4.1 個案公司—H公司 49
4.1.1. 維修零件供應鏈 52
4.1.2. 存貨管理政策 54
4.1.3. 零件供應端 58
4.1.3.1. 供應端預測需求之變異 58
4.1.3.2. 備料供應商 61
4.1.4. 小結 63
4.2 維修零件分群演算 64
4.2.1 資料整理與標準化 64
4.2.2 FCM分群演算法 65
4.2.3 最適分群數決定 66
4.3 訂購參數決定 69
4.3.1 群組內平均前置時間與樣本標準差 70
4.3.2 最適訂購週期 70
4.3.2.1 情境限制說明 70
4.3.2.2 個別零件最適訂購週期求解 72
4.3.3 最適訂購策略 75
4.4 模擬驗證 76
4.4.1 模擬情境說明 76
4.4.2 模擬求解實例 79
4.5 敏感度分析 83
4.5.1 改變緊急空運與海運費率差額 83
4.5.2 改變存貨持有成本 84
4.6 小結 85

第五章 結論與建議 87
5.1 研究結論 87
5.2 研究貢獻 88
5.3 研究限制 89
5.4 未來研究方向 90

參考文獻 91
網路資源 93
附錄A：XB Index 94
附錄B：預期安全存貨（S） 95
附錄C：標準化期望缺貨個數 ( ) 96
附錄D：各群聚內最適化訂購週期（G2~G7） 100

[1] 林姿依，建立適合顧客關係管理之模糊分群模型—以汽車維修服務為例，台灣大學商學研究所，碩士論文，台北，2007。
[2] 張保隆、陳文賢、蔣明晃等，生產管理（三版），華泰文化，台北，2006。
[3] 黃明顯，建構汽車售後服務零件供應鏈創新策略之方法與實證研究—以和泰汽車為例，東吳大學企業管理學系，碩士論文，台北，2005。
[4] 盧銘勳，變動前置時間之存貨模型之建立，逢甲大學工業工程與系統管理學系，碩士論文，台中，2005。
[5] Ankerst, M., Breunig, M., Kriegel, H. P., & Sander, J., Optics: Ordering Points to Identify the Clustering Structure, In Proc. 1999 ACM-SIGMOD Int. Conf. Management of Data (SIGMOD’99), Philadelphia, PA, June, 1999, 49-60.
[6] Ballou, R.H, Business Logistics/Supply Chain Management, 5th edition, Person Education, Inc, Upper Saddle River, N.J., 2004.
[7] Bezdek, J.C., Pattern Recognition with Fuzzy Objective Function Algorithms. New York: Plenum Press, 1981.
[8] Cakanyildirim, M., Bookbinder J. H. & Gerchak Y., Continuous Review Inventory Models Where Random Lead Time Depends on Lot Size and Reserved Capacity, International Journal Production Economics, Vol.68, 2000, 217-228.
[9] Cohen, M. A., Kleindorfer P. R. & Lee H. L., Near-Optimal Service Constrained Stocking Policies for Spare Parts, Operations Research, Vol.37, No.1, 1989, 104-117.

[10] Cooper, M. C., Lambert, D. M. & Pagh, J. D., Supply Chain Management: More Than a New Name for Logistics, The International Journal of Logistics Management, Vol 8, Issue 1, 1997, 1-14.
[11] Dunn, J. C., Well-separated Clusters and the Optimal Fuzzy Partitions. Journal of Cybernetics, Vol. 4, 1954.
[12] Feller, W., An Introduction to Probability Theory Its Applications Vol.1., Wiley, New York, 1960.
[13] Kaufman, L., & Rousseeuw, P.J., Finding Groups in Data: An Introduction to Cluster Analysis. John Wiley & Sons, 1990.
[14] MacQueen, J.B., Some Methods for Classification and Analysis of Multivariate Observation, In Proc. 5th Berkeley Symp. Math. Statistics, 1967, 287-297.
[15] Ng, R. & Han, J., Efficient and Effective clustering Method for Spatial Data Mining, In Proc. 1994 Int. Conf. Very Large Databases (VLDB’94), Santiago, Chile, Sept, 1994, 144-155.
[16] PAN, J.C. et al. , Inventory Models with Fixed and Variable Lead Time Crash Costs Considerations. Journal of the Operational Research Society Vol.53, Issue 9, 2002.
[17] Silver, E.A., Peterson R. , Decision Systems for Inventory Management and Production Planning, John Wiley, New York, 1986.
[18] Simchi-Levi, Kaminsky, P. & Simech-Levi, E. ,Designing and Managing the Supply Chain : Concepts, Strategies, and Case Studies, 2nd, McGraw-Hill Companies, Inc. New York, 2003.

[19] Tersine, R. J., Principles of Inventory and Materials Management, PTR Prentice-Hall, Inc.
[20] Weiss, S. M., and Indurkhya, N., Predictive Data Mining: A Practical Guide. Morgan Kaufmann, CA, 1998.
[21] Xie, X.L., and Beni, G., A Validity Measure for Fuzzy Clustering. IEEE Transaction on Pattern and Machine Intelligence, Vol.13, Issue 8, 1991, 814-847.