整合顧客需求與預測以提升存貨管理

Integrating Customer Demand and Forecast to Enhance Inventory Management

呂尚霖(Shang-lin LU)

指導教授 ： 黃惠民

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2017年)

https://doi.org/10.6840/cycu201700069

摘要

近年來，供應鏈的活動影響了社會和環境，為了避免在供應鏈下游提出需求後發生缺貨的問題，上游端會建立庫存系統來確保能供應下游的需求變動，因此存補貨管理無疑是供應鏈管理中重要其中一個環節。相較於傳統補貨策略需要較多且複雜的參數運算求出最佳解，本研究利用限制理論(Theory of Constraints, TOC)搭配需求拉動補貨(Demand-Pull)與緩衝管理(Buffer Management)並輔以指數加權移動平均法(Exponentially Weighted Moving Average, EWMA)，只需要參考顧客即時需求資訊與需求趨勢便能做出補貨的策略。
本研究主要考量的產品為具有生命週期短、需求變動大特性之半導體封裝產業，因此利用指數加權移動平均法(Exponentially Weighted Moving Average, EWMA)整合顧客提供之實際需求預測(Forecast)與顧客實際需求量(Demand)掌握顧客需求變動趨勢。並觀察在庫量水位於之緩衝區域搭配EWMA指標正、負向需求趨勢管理緩衝與調整訂購量。藉由EWMA特性能反映時間內曲線小幅度變化，能快速反應改善補貨策略。本研究以國內某封裝廠實際案例分析，並依照不同產品需求模式提供權重，在需求變異大(小)、預測變異大(小)、需求與預測差距大(小)之產品特性中，評估出本研究方法適用性。統整分析後結果顯示本研究確實能適用且加強限制理論補貨策略之應用，達到整體庫存量降低之效果。

關鍵字：供應鏈管理、限制理論、緩衝管理、需求拉動補貨、指數加權移動平均

關鍵字

供應鏈管理、限制理論、緩衝管理、需求拉動補貨、指數加權移動平均

並列摘要

In recent years, supply chain activities play an important role in society and environment. In order to avoid the supply chain downstream demand running out of stock problems, the upstream end of the inventory system is controlled to monitor the downstream demand changes. Therefore, the replenishment management within the supply chain is very important. Compared with the traditional replenishment strategy, which requires more complex parameter operations to find the optimal solution, this study uses the Theory of Constraints (TOC) and demand-pull (Buffer Management). Supplemented by an Exponentially Weighted Moving Average (EWMA) strategy, one can replicate and refer to the customer's immediate demand information and demand trends.
In this study, we consider the semiconductor packaging industry which is characterized by short life cycle and large demand fluctuation. Therefore, we use the Exponentially Weighted Moving Average (EWMA) to integrate customer demand forecast with customer actual demand to monitor the trend of changes in customer demand. We then observe in the buffer zone the amount of inventory in the zone. EWMA will indicate positive and negative demand trend; management can then adjust the order quantity. EWMA features can reflect the small changes in the trend and quickly adjust the replenishment strategy. In this study, the actual data of a domestic packaging plant is used. Different weights and some parameter variations are done for different demand characteristics of the products. They are large (small) demand variations, large (small) forecast variations, large (small) demand and forecast variations. The results of our model in this study not only strengthen the theory of behind the replenishment policy, it also reduces the overall inventory cost.

Keywords：Supply chain management, theory of constraints, buffer management, demand-pull strategy, exponentially weighted moving average

並列關鍵字

Supply chain management, theory of constraints, buffer management, demand-pull strategy, exponentially weighted moving average

參考文獻

63. 王晃三、顏貽楨，2002，限制理論的問題管理模式，中華民國品質學會第 38 屆年會暨第 8 屆全國品質管理研討會。

5. Akhundjanov, S. B., & Pascual, F. (2015). Moving range EWMA control charts for monitoring the Weibull shape parameter. Journal of Statistical Computation and Simulation, 85(9), 10.1080/00949655.2014.907574

7. Chopra, S. and Meindl, P. (2007). “Supply Chain Management: Strategy”, Planning and Operation, Vol. 3.

9. Cooper, M. C., Lambert, D. M. and Pagh, J. D. (1998). “Supply Chain Management: Implementation Issues & Research Opportunities”. The International Journal of Logistics Management, Vol. 9(2), pp.1-19.

10. Chang, Y. C., Chang, K. H., & Sun, W. C. (2015). Enhancement of Inventory Management for the Wafer Manufacturing Industry by Combining Market Demand Forecast and Demand-Pull Replenishment. Journal of Testing and Evaluation, 43(4), 948-963.