降低訂購成本之供應鏈存貨模式

Abstract

面對日益普及的網路科技，企業間運用電子商務改善交易作業的需求日趨顯著，企業對企業電子商務(business to business electronic commerce, B2B EC)不僅大幅提昇上下游交易的效率，也強化供應鏈交易夥伴間的合作關係，本文目的在於探討供應鏈改善交易作業對於上下游存貨決策與存貨成本的影響。有關供應鏈存貨決策之探討，理論上多見於聯合存貨模式(integrated inventory model)之研究，聯合存貨模式是基於交易夥伴存貨總成本之最適化，整合性地制定各成員之存貨決策。而就存貨理論，訂購成本(ordering cost)代表交易雙方為完成一筆交易如：詢價、報價、訂貨、登錄訂單、備貨、出貨、開立發票、驗貨及付款等所須付出的作業成本，其不僅反映交易流程的效率亦攸關存貨決策的擬定。傳統上，訂購成本被視為固定且不可控制的參數，然實際上，訂購成本可透過交易流程合理化與資訊科技的運用而獲得改善，即訂購成本可藉由額外投入的金額加以降低，近年來，已有學者針對企業投資改善交易作業的情況，探討投資金額與存貨成本間交互損益(trade-off)的關係，分析降低訂購成本下之存貨決策，然目前有關可調降訂購成本之存貨模式，多是針對個別企業存貨系統，以內部作業改善決策觀點視之，並不適合供應鏈包含上下游存貨系統，涵蓋多個企業決策的情形。據此，本文以聯合存貨模式為基礎，納入可調降訂購成本，針對幾種基本的供應鏈存貨系統，構建量化模式，藉以分析降低訂購成本對於供應鏈存貨決策與存貨成本之影響。 為達成研究目的，首先針對單一供應商與單一買方供應鏈發展一聯合存貨模式，其延伸Goyal & Srinivasan (1992)的研究，涵蓋允許買方延遲交貨的情況，以增進模式的一般性，並分別就聯合決策、買方主導(buyer dominant)與供應商主導(vendor dominant)等三種情況推導買、賣方之最適批量；此外，由數據實驗的結果發現：相較於買方主導與供應商主導情況，聯合決策下供應鏈單位時間存貨總成本為最小，而供應商主導所代表的推式(push)系統，其單位時間存貨總成本經常高過其餘兩種情況。其後依據Porteus (1985)的理論，假設訂購成本為單位時間投入金額的一般性遞減(general decreasing)函數，提出三個可調降訂購成本之聯合存貨模式：首先是針對單一供應商與單一買方供應鏈，乃基於前述聯合存貨模式並納入降低訂購成本決策，由模式分析的結果發現買方的訂購批量與延遲交貨量將隨訂購成本的降低而減少。其次是針對單一供應商與多買方供應鏈設置供應商管理存貨(vendor-managed inventory, VMI)系統的情況，以Banerjee & Banerjee (1992)的批對批共同週期(common cycle)模式為基礎，加入可調降訂購成本與買方延遲交貨決策，由分析的結果顯示：在降低訂購成本下，上下游的庫存水準皆會下降，買方的延遲交貨量亦隨之減少。最後是針對單一製造商、單一配銷商與多零售商所組成之供應鏈，由模式分析的結果顯示，降低訂購成本可以減少配銷商的訂購批量與配送至各零售商的出貨量，使得供應鏈中下游的庫存負擔下降，同時各零售商的延遲交貨量亦將減少。上述三個模式，除了就一般性訂購成本函數分析批量的性質，並分別就模式於指數訂購成本函數下的特性進行探討，此外，亦藉由數據釋例分析決策與單位時間系統總成本對於參數變動之敏感度。 有關資訊科技改善交易作業的效益諸如減少人力需求、減少資料傳遞錯誤、縮短供貨前置時間、減少庫存、快速反應市場需求以及強化買賣方合作關係等，由本研究的結果顯示：單就存貨成本的節省即可能確認調降訂購成本的經濟可行性。雖然提昇交易效率、降低訂購成本已成為企業減少存貨與物流成本的重要策略，然理論上可供分析、解釋降低訂購成本下供應鏈存貨決策之量化模式仍不多見，期藉由本文做為發展此類研究之起點。

Forming strategic alliances and utilizing information technologies in logistics are increasingly important for firms to gain competitive advantages. A popular application of information technology is the business to business electronic commerce (B2B EC) which provides numerous benefits such as the reductions in transaction time, data errors, and inventory costs, and results in better vendor-buyer coordination and more streamlining supply chains. More and more firms have devoted much time and capital to the reduction of ordering costs through the application of information technologies, but there have been few models available to investigate and facilitate such efforts. The objective of this dissertation is to provide models to explore the impact of investment in ordering cost reduction on lot-size decisions for integrated inventory systems in which the trading parties have formed a strategic alliance for profit sharing. The study initially focused on a supply chain consisting of a single vendor and a single buyer in a deterministic environment. An integrated inventory model, which extends Goyal & Srinivasan’s work (1992) by incorporating the buyer’s backorders decision, is proposed to derive the optimal vendor and buyer lot-size decisions with their joint total cost minimized. The optimal lot sizes and total system cost under the vendor-buyer coordination policy are compared with that under the buyer-dominant policy and the vendor-dominant policy. According to the work of Porteus (1985), ordering costs can be reduced by extra capital expenditures and are assumed to be general decreasing functions of capital expenditure per unit time. Three models are presented next to investigate the trade-off between the investment in ordering-cost reductions and the operating costs of various integrated inventory systems. The first model extends the preceding integrated inventory model for a single vendor and a single buyer inventory system by incorporating the ordering cost reduction decision. The second one is proposed for a single vendor and multiple buyers inventory system in a deterministic environment. It extends the common cycle model of Banerjee & Banerjee (1992) by involving both the buyers’ backorders and the ordering cost reduction decisions. The third model is developed for an integrated production-distribution system consisting of a single manufacturer, a single distributor, and multiple retailers. For each model, the optimal lot sizes and capital expenditures in ordering-cost reductions for minimizing the total system cost are derived, and the application of models are further illustrated by exponential ordering cost functions. It has been shown that the lot sizes of buyers will decrease once the supply chain invests in ordering-cost reductions, which confirms the trend of smaller-lot-size deliveries. Numerical experiments and sensitivity analyses are also preformed for the exponential case for each model, which give managerial implications of the effects of parameters on decisions.