سیاست های موجودی بهینه در زنجیره تامین غیر متمرکز
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|20608||2010||7 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : International Journal of Production Economics, Volume 128, Issue 1, November 2010, Pages 303–309
This paper studies the inventory behavior of autonomous and self-serving firms in a decentralized retailer–manufacturer serial supply chain. We first analytically characterize the optimal inventory policies for each firm both with and without information sharing, revealing an inherent simple structure of optimal inventory behavior: replenishments are triggered by ultimate customer demand directly. We then consider various extensions: (1) N-stage serial systems, (2) batch ordering policies, (3) fixed setup costs, and (4) Markovian customer demand.
This paper studies inventory behavior in a market economy, in which firms first are autonomous and business is pulled by demand. Every firm aims to maximize its own performance, while working very hard to satisfy its customers’ demand. Activities of sales and operations are triggered by downstream orders. In other words, firms optimize their own individual performance based on what is happening downstream where buyers follow some rational decision-making policies for themselves. We first pause here to post an important question to industry: what is real demand? In practice, orders are often successively passed upstream—the so-called demand evolution (Li, 2008). Facing the customer demand a facility purchases from its supplier based on some popular replenishment policies such as (s, S) or (R, nQ) policies. The order is demand information for its supplier. In other words, the replenishment policies transform facility’s demand information (i.e., customer demand) into its supplier’s demand information (i.e., facility’s orders). This phenomenon propagates along the whole supply chain from downstream to upstream. Firms often use their direct downstream orders as real demand to determine their production and inventory flows. However, these orders can be riskier and illusive, due to self-interest of downstream customers (e.g., minimizing their own costs) ( Li, 2008). The paper promotes the concept of ultimate demand, showing that the real demand actually is ultimate customer demand even with the focal point of each individual self-serving firm. Orders in between supply chain members are “evolved” ultimate demand. Replenishments should be triggered by ultimate demand directly and exactly. In this way we promote a straightforward approach to manage multi-stage decentralized supply chains where every firm optimizes its own performance. Our contribution is that we explicitly characterize evolved demand that is compared to original demand; we then derive optimal policies based on demand characteristics and obtain information value based on the comparison. We show that all orders in between supply chain members can be traced back to the origin: ultimate demand.
نتیجه گیری انگلیسی
This paper reveals an inherent simple structure of optimal inventory policies for self-serving firms in decentralized supply chains: the optimal replenishments are triggered by ultimate demand. In this way we extend literature results of system performance in centralized supply chains to where firms are self-serving. This ultimate-demand concept is important from the systems aspect since the real demand is ultimate demand, rather than orders in between supply chain members. All orders should be treated as evolved ultimate demand. Our plain idea insures that all firms work on the same and true base: ultimate demand. In this way we promote a simple approach to manage decentralized supply chains where every firm only optimizes its own performance separately. We also present the value of demand information. Without the knowledge of customer demand, the upstream supplier has to use immediate downstream orders that are riskier than ultimate customer demand. Having the knowledge changes the supplier’s ordering behavior, and lowers his cost and the system cost. Our results thus answer two information sharing questions: whether or not to pass consumer demand data (e.g., point-of-sale data) to suppliers and how to use them. Our approach can be applied to non-stochastic demand and other business problems. The idea is that the firm can more accurately anticipate its customers’ behavior and thus satisfy them better after knowing their states.