消費性電子創新產品供應鏈存貨系統之方法論

Methodologies for solving supply chain inventory systems for consumer electronic innovative products

黃彥登(Yen-Deng Huang)

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/博士(2015年)

https://doi.org/10.6840/cycu201500060

摘要

摘要
近年來由於全球市場的競爭日趨激烈，已使得消費性電子產品的邊際利潤急遽下降，進而迫使企業不得不去尋求有效地降低總成本的方式，以提升企業的自我競爭力與整體利潤。而且消費性電子產品本身也具有短生命週期和需求的不確定性的因素，這將會導致定價及管理策略是非常難以處理，尤其是針對3C產品（如智慧型手機，iPhone 6和6 Plus, 2014）、遊戲機、電子產品及零部件加工產業等。因此企業在制定製造或服務策略時，不再是一個單獨運作的個體，而是講求企業結盟的夥伴關係(Partnership)，藉由上、下游間的垂直整合來提升供應鏈整體的效益。因此有效的運用協同合作讓企業間的資源分享及資訊透明化，將是供應鏈協調成功的重點。且對於工業界和學術界而言，整合消費性電子產品供應鏈等相關的研究議題在近年來已逐漸受到重視。

再者，為了達成快速回應(Quick Response)以滿足顧客需求為目標，接駁式轉運(Cross Docking)則為另一項可採用的物流策略。世界知名廠商Wal-Mart因採用供應商管理庫存(Vendor Management Inventory, VMI)與接駁式轉運技巧，有效降低存貨留在倉庫的時間，進而擁有壓低商品售價及降低營運成本。

因此，本研究探討整合消費性電子產品供應鏈之存(補)貨模式，冀望能創造一個雙贏的局面，使得供應鏈中的所有成員皆能獲利。其中，在供應鏈模型中考量不同類型的協調機制，雙倉存貨理論及接駁式轉運來控制的需求的不確定性，並進一步追求整體利潤最大化（或成本最小化）之隨機最佳化問題。本研究將提出一套有效的回溯粒子群演算法(PSORA Algorithm)來求得整合消費性電子產品供應鏈之存(補)貨模式的最佳解。回溯粒子群演算法是以回溯近似法(Retrospective Approximation, RA)的為基礎架構；藉由遞增模擬樣本數來反覆地解決一系列的樣本路徑(Sample path)近似問題並透過粒子群演算法(Particle Swarm Optimization, PSO)找尋每一個樣本路徑問題之最佳解。當樣本數遞增到無限大時，此樣本路徑即等同於真實的目標函數。之後透過數值範利和敏感度分析進行驗證及提供各參數對模式之影響。最後，本研究的研究成果將提供決策者在管理與協調消費電子創新產品供應鏈的整合能力。

關鍵字

接駁式轉運 ； 雙倉存貨 ； 需求不確定 ； 供應鏈協商整合 ； 消費性電子創新商品 ； 回溯近似法 ； 粒子群演算法 ； 樣本路徑近似問題

並列摘要

ABSTRACT
Due to the competition of innovative products in the global market, products have decreasing the profit margin. Consequently, enterprises have to seek ways to cut the cost and enhance their competitiveness. Moreover, due to the short-life cycle and demand uncertainty of the consumer electronic innovative products, pricing and management strategies are extremely difficult to handle; it is especially true for 3C products (e.g., smart mobile phone, iPhone 6 and 6 Plus sold in 2014), game consoles, electronics and components processing industries in Taiwan. Hence, an emphasis on supply chain integration and coordination through resource and information sharing has been an important issue and becomes an improvement for all enterprises in recent years. These researches considering the above factors have received considerable attention from industries and academia.

In order to improve customer satisfactions, cross-docking is practiced by businesses. Wal-Mart, a company famous for using Vendor Managed Inventory (VMI) and cross docking, has used these techniques to decrease storage time and increase speed in products logistics. With these strategies, operating costs and product prices can be reduced significantly.

Our research focuses on creating a win-win situation where both consumer satisfactions and businesses profits for all members in the supply chain will increase. The integrated inventory (replenishment) supply chain model developed for innovation products uses different types of coordination mechanisms, two warehouses and cross-docking technique to control response time and demand uncertainty and to further pursue the overall profit maximization (or cost minimization). Based on this, an optimum strategy can be derived for this kind of stochastic optimization problem. An efficient/effective artificial intelligent heuristic algorithm (Particle Swarm Optimization Retrospective Approximation, PSORA) is developed to solve the complex problem; a numerical example and sensitivity analysis are carried out to validate the model. PSORA uses the framework of RA (Retrospective Approximation) to iteratively solve a sequence of sample path approximation problems with increasing sample sizes, where each sample path approximation problem is solved by the PSO algorithm. When the sample size goes to be infinite, the PSO algorithm solves the sample path approximation problem which is identical to the real objective function. Finally, the model can be used to assist in decision making for managing consumer electronic innovative 3C products. The model developed will provide marginal insights to decision-makers in enhancing the ability of the enterprises to manage the consumer electronic innovative products supply chain.

並列關鍵字

Demand uncertainty ； Two warehouses ； Cross-docking ； Particle swarm optimization ； Retrospective approximation ； Sample path approximation problem ； Consumer electronic innovative products ； Supply chain integration

參考文獻

Philippe, G., and Goldberg, D. E., 1998, “Population sizing for optimum sampling with genetic”, In Proceedings of the Third Annual Genetic Programming Conference, Madison, WI, July 22-25, 1998. Philippe Gigu ere Aeronautical and Astronautical Engineering Department.

Abdul-Jalbar, B., Segerstedt, A., Sicilia, J., and Nilsson, A., 2010, “A new heuristic to solve the one-warehouse N-retailer problem”, Computers & Operations Research, Vol. 37(2), pp. 265-272.

Agarwal, G., Barari, S., and Tiwari, M. K., 2012, “A PSO-based optimum consumer incentive policy for WEEE incorporating reliability of components”, International Journal of Production Research, Vol. 50(16), pp. 4372-4380.

Agrawal, S., and Banerjee, S., 2013, “Inventory model with deteriorating items, ramp-type demand and partially backlogged shortages for a two warehouse system”, Applied Mathematical Modelling, Vol. 37, pp. 8912-8929.

Agustina, D., Lee, C. K. M., and Piplani, R., 2014, “Vehicle scheduling and routing at a cross docking center for food supply chains”, International Journal of Production Economics, Vol. 152, pp. 29-41.