過去，為因應全球景氣發展，特殊合金鋼產業雖極力擴大產能仍處於供不應求的情況，但是一場金融風暴改變此情況，買家訂單瞬間蒸發，因此產品無法順利銷售，導致倉庫擺放堆積如山的成品。雖然歷經數年後，全球需求逐漸回到過去水平，甚至有逐漸成長趨勢，國內企業仍必須針對存貨管理作進一步的管理。
本研究透過一個實際案例來進行存貨管理之研究，個案公司使用MTS/MTO生產系統之產品有存貨過多問題，希望求得一個適當的需求預測方法，計算出與未來需求相近的預測值，找出與需求預測之相關因素，針對此因素來建構本研究的預測模式，使用移動平均法、加權動移動平均法、指數平滑法及最小平方法作為預測客戶需求之方法，模式建構完成後，針對模式進行運算與分析。上述即為本研究的模式建構部分，當模式建構完成後，將進行專家訪談，請專家從五種參數選擇，最終將其參數調整預測值，冀望能夠調整出更符合現實需求之預測，讓決策者可以利用此模式來進行需求預測之決策。
經分析結果發現，加權移動平均法所預測之結果最符合客戶需求，並採用問卷方式訪問個案企業中的三位專家，挑選出外部參數-市場潛力。依據不同產品別給予不同的外部參數，2010年的預測值整體來看，兩者差異不大；2011年的預測值整體來看，透過調整的預測值可將庫存差異在調降9%，因此未來可考慮加入專家所提出外部參數對預測值進行調整，最終以該預測值作為半成品庫存數量之依據。

The purpose of this study is to establish an inventory management system suitable for the special alloy steel industry which meets existing requirements while maintaining the inventory of semi-finished products at the lowest possible level. Therefore, this study utilized methods of moving average, weighted moving average, exponential smoothing, and least squares to predict customer demand and, using comparison, determine which method also offers the lowest level of inventory. The expected results were adjusted based on parameters selected with the help of interviews with specialists, in the hope that the system meets customer demand and provides a reference for semi-finished product management strategies in the future.
The results showed that the prediction method most suitable for the subject industry was the weighted moving average method, which yielded prediction values that were closer to actual customer demand and was able to effectively decrease the amount of semi-finished products in the inventory - more effectively than the inventory management method previously deployed by the subject company. Also, analysis showed that the opinions from specialists had a certain level of influence. Therefore, when conducting demand prediction, specialist opinions must be taken into consideration.

Adan, I., & van der Wal, J. (1998). Combining make to order and make to stock. OR Spektrum, 20(2), 73-81.
Anthony, R. N. (1965). Planning and control systems: a framework for analysis: Division of Research, Graduate School of Business Administration, Harvard University.
Heizer, J., Render, B. (2010 ). 作業管理原理（賴意銓、張皓維譯）. 雙葉書廊.
Jonsson, P., & Mattsson, S. A. (2013). The value of sharing planning information in supply chains. International Journal of Physical Distribution & Logistics Management, 43(4), 282-299. doi: 10.1108/ijpdlm.07-2012-0204
Kiris, S. (2013). Multi-Criteria Inventory Classification by Using a Fuzzy Analytic Network Process (ANP) Approach. Informatica, 24(2), 199-217.
Olhager, J. (2003). Strategic positioning of the order penetration point. International Journal of Production Economics, 85(3), 319-329. doi: 10.1016/s0925-5273(03)00119-1
Porter, K., Little, D., Peck, M., & Rollins, R. (1999). Manufacturing classifications: relationships with production control systems. Integrated Manufacturing Systems, 10(4), 189-199.
Rafiei, H., & Rabbani, M. (2011). Order partitioning and Order Penetration Point location in hybrid Make-To-Stock/Make-To-Order production contexts Computers & Industrial Engineering (Vol. 61, pp. 550-560).
Rafiei, H., & Rabbani, M. (2012). Capacity coordination in hybrid make-to-stock/make-to-order production environments. International Journal of Production Research, 50(3), 773-789. doi: 10.1080/00207543.2010.543174
Rafiei, H., Rabbani, M., & Alimardani, M. (2013). Novel bi-level hierarchical production planning in hybrid MTS/MTO production contexts. International Journal of Production Research, 51(5), 1331-1346. doi: 10.1080/00207543.2012.661089
Soman, C. A., van Donk, D. P., & Gaalman, G. (2004). Combined make-to-order and make-to-stock in a food production system. International Journal of Production Economics, 90(2), 223-235. doi: 10.1016/s0925-5273(02)00376-6
Tersine, R. J. (1982). Principles of inventory and materials management.
Wang, F., R. Piplani, Y. Roland, and E. Lee. (2011). Development of an Optimal Decision Policy for MTS-MTO System.
Watts, C. A., Hahn, C.K., and Sohn, B.K. (1994). Monitoring the Performance of a Reorder Point System: A Control Chart Approach. International Journal ofOperations & Production Management, Vol. 14, No. 12, pp. 51-61. doi: 10.1108/01443579410053211
于宗先. (1972). 經濟預測: 中央研究院經濟研究所.
林清河. (2010). 工業工程與管理. 華泰文化.
徐依筠. (2010). 考量回廠率下零組件需求預測與備貨模式之研究—以汽車業為例.
黃錫鴻. (2008). 應用灰色理論預測半導體設備消耗性零件需求量. 國立中央大學.
劉文海. (2012). 台灣特殊鋼產業發展現況. 金屬中心產業研究組.
劉振宏. (2010). 具效率性的指數平滑法應用於OEM庫存預測之研究.
謝秀圓. (2012). 不同藥品耗用類型預測暨庫存管理之研究-以某醫學中心為例.
關智峰. (2009). 維修零件最終訂購問題之需求預測模式.