供應鏈滾動預測與訂單策略之模擬研究

A Simulation on Supply Chain Rolling Forecast and Inventory Strategy Research

謝依潔(I-Chien Hsieh)

指導教授 ： 范錚強

國立中央大學/管理學院/資訊管理學系/碩士(2007年)

摘要

本研究觀察到目前台灣許多產業供應商交貨時間被高度壓縮，因此必須倚賴滾動預測的方式來進行事先生產或備料。但在目前學界的研究中，卻缺少對滾動預測的研究。這引發本研究針對滾動預測，利用電腦模擬方法來進行這項實驗研究。首先，本研究先對滾動預測現象及案例進行探討。接著，本研究建立一個具有單一供應商及單一零售商的二階層供應鏈。研究假設其顧客訂購要求之交期較上游供應商前置時間短，經由預測訂單的資訊分享，且考慮本身成本最小化的策略。這主要是來探討前置時間、安全庫存以及存缺貨成本比例的訂貨調整策略對供應鏈績效的影響。各種實驗的情境包括預測高估、低估及一般無偏差的情境，配合誤差平均、誤差平均標準差以及誤差平方根三種訂貨調整策略。本研究結論為，在不同的預測誤差或預測偏誤下，廠商如能選擇合適的訂貨策略以及縮短上游廠商的前置時間，加上對產業別的調整，可為提高供應鏈績效的最好方法。

關鍵字

預測偏誤 ； 預測誤差 ； 滾動預測 ； 供應鏈

並列摘要

Based on the current phenomena of the Taiwan industries, it is apparent that the order delivery time has been severely compressed to meet customer demand. A straight forward solution may end up with serious inventory problems. In order to solve the problem, “rolling forecast” has been widely employed to cope with the problem, such that materials can be prepared in advance. However, there is a dearth in academic research on this issue. This research started out with an investigation on the issue of rolling forecast applications in the Taiwan industry. Subsequently, the computer simulation method was used to study the problem on hand. A two echelons supply chain with one focal supplier and one customer, where deliver time is shorter than upstream supplier lead time was assumed. Through order information sharing and minimizing total inventory cost, the simulation experiments investigated the impact of safety stock, lead time and different inventory and out of stock cost proportion.
Many scenarios were included in the simulation study, including forecast bias forecast error, out of stock penalty, and inventory replenishment strategies. In conclusion, we found out that in different forecast error and forecast bias settings, the companies have to choose adaptable strategy and shorten the upstream lead time. Moreover, different industries have to adjust their replenishment strategies to improve their supply chain performance.

並列關鍵字

Forecast Bias. ； Forecast Error ； Rolling Forecast ； Supply Chain

參考文獻

1. 林則孟 (2006)，生產計劃與管理，華泰文化事業股份有限公司：台北市。

4. 郭瑞祥，蔣明晃 ，陳亞男，劉基全 (2004)，「多廠生產規劃之供應鏈決策支援模式」， 臺灣大學管理論叢，15(1), pp. 49-74。

2. 張怡筠 (2006)，「CPFR中滾動式預測的關鍵因子探討」，國立清華大學資訊系統與應用研究所碩士論文。

1. Aviv, Y. (2001), “The Effect of Collaborative Forecasting on Supply Chain Performance,” Management Science, 47(10), pp. 1326-1343.

2. Biggs, J., and Campion, W. (1982), “The Effect and Cost of Forecast Error Bias for Multiple-Stage Production-Inventory Systems,” Decision Sciences, 13(4), pp. 570-584.