多晶矽原料存貨策略之研究-以太陽能晶圓廠為例

Poly-Silicon Inventory Strategy – A Case Study of Solar Energy Fab

范谷銓(Ku-Chuan Fan)

指導教授 ： 饒忻

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2012年)

https://doi.org/10.6840/cycu201200875

摘要

多晶矽原料約佔多晶矽晶片總生產費用之59%，因此對於太陽能矽晶片生產商來說，多晶矽之費用對其銷售價格及生產成本影響甚重。另外，多晶矽原料價格會依市場趨勢經常性地變化；當多晶矽原料市場價格大幅上揚時，其多晶矽原料之安全庫存也將成為資金積壓的主要成本，所以探討有效之存貨策略，以降低成本是必要的。
本研究以太陽能晶圓廠D公司為例進行個案研究，建構(訂購點，訂購量：R，Q)模型。在求取最佳訂購點部份是由再訂購點(Reorder Point：ROP)模型延伸而來，利用4週移動平均法預估下期矽原料平均使用量與安全庫存量，計算ROP來控制矽原料庫存量。在訂購量上則是依最小運輸成本方式選擇大於或等於ROP的最小矽原料購買量。為配合移動平均法的計算，在每次庫存量低於ROP進行補貨程序時，此(R，Q)模型需重新計算，以擷取最貼近趨勢的新資料。本研究並與現況、ROP、定期訂購等三種模型進行比較，結果顯示本(R，Q)模型在不降低服務水準的前提下，有較低的矽原料庫存成本，可以提昇企業競爭力。

關鍵字

多晶矽 ； 移動平均法 ； 再訂購點 ； 庫存

並列摘要

Polycrystalline silicon charges nearly 59% in the overall production cost of multi-crystalline solar wafer, thus for a solar wafer maker, the selling price and production cost are heavily influenced by the cost of polycrystalline silicon. In addition, the price of polycrystalline silicon depends on the market trend and changes frequently. When the market price of polycrystalline silicon largely increases, the safety stock of poly-silicon will be the main cost of capital accumulated. Therefore, a study of effective inventory strategy is essential in order to reduce cost.
This research uses a solar energy fab of D company as a case study and sets up an (Order point, Order quantity; R, Q) model to solve high inventory cost problem. A reorder point model will be used to find the optimal order point by using the 4-week moving average method in order to estimate the demand and safety stock of polycrystalline silicon for the next period. With regard to the order quantity, we will choose a quantity with the smallest transportation cost, which is is larger than or equal to the reorder point quantity. This (R, Q) model has to renew while the inventory is less than the reorder point quantity. In addition, this study compares with the current practice, ROP model and regular order system, and the result indicates that the proposed model has the lowest inventory cost of polycrystalline silicon without a sacrifice of service level, which can enhance the competitiveness of company D.

並列關鍵字

Inventory ； Reorder point ； Polycrystalline silicon ； Moving average method

參考文獻

白賜清 (1998)，品質管制之統計方法，第五版，中華民國品質學會。

Chang, P.T., Chang, C.H., and Jong, T.P. (2001), "A Dynamic Unit Cost Based Economic Production Quantity Model with Fuzzy JIT Concept ", Journal of the Chinese Institute Of Industrial Engineers, Vol. 18, pp. 9-20.

Huang, Y.C. (2001), "A Study On Optimal Ordering And Pricing Policy Under Probabilistic Demand And Quantity Discount For Extended Newsboy Problem ,"Journal Of The Chinese Institute Of Industrial Engineers,Vol.18, No.6,pp.43-52.

Kim, J.S. and Benton W.C. (1993), "Lot size depend en t lead time in a Q, R inventory system "International Journal of Production Research, Vol 33,pp.41-58.

Ben-Daya, M. And Raouf A. (1994), "Inventory models involving lead time as decision variable "Journal of the Operational Research Society, Vol 45,pp.579-582.