外文翻譯---供應鏈管理環(huán)境下的庫存優(yōu)化

1、<p><b>　　附錄一 英文原文</b></p><p>　　INVENTORY OPTIMUM BASED ON SUPPLY CHAIN MANAGEMENT</p><p>　　YUN Jun YAN Bing ZHAO Yuwei</p><p>　　College of Management of W

2、uhan University of Technology Hubei Wuhan</p><p>　　Abstract: Because the optimized inventory in traditional supply chain model has poor information, it becomes more difficult to obtain optimal solution com

3、plying with the practical requirements during finding, solutions to supply chain patterns. This article is intended to analyze the operational mechanism of optimized inventory in both traditional enterprises and supply c

4、hain management. Also，this article put forward to improve traditional inventory patterns with the aid of multiple-layer BP neu</p><p>　　Key Words: Supply Chain, SCM, Inventory, BP Neural Network, Optimized&l

5、t;/p><p>　　1.INTRODUCTION</p><p>　　Supply chain Management (SCM for short) is a hot topic today. The term supply chain comes from a picture of how organizations are linked together as viewed from a

6、 particular company. The idea of SCM is to apply a total systems approach to managing the entire flow of information, materials, and services from raw materials suppliers through factories and warehouses to the end custo

7、mers . Successful SCM requires an integration of series activities into a seamless process. However, there must be som</p><p>　　Artificial neural network have the ability to learn by itself and multi-mapping

8、, and it can explore complicate system escaping to make complicate models. In the artificial neural network models, the information hides in the network made by linked-neuron, and it can deal with multiple quantitative r

9、elationships. Namely, the ANN is a massively parallel computational model, and it has characterizes : </p><p>　　·Great degree of robustness and fault tolerance;</p><p>　　·Ready to deal

10、 with problems associated with general nonlinear systems;</p><p>　　·Biophysical implications.</p><p>　　So ANN is a good analysis tool for nonlinear problem. This paper will put forward to i

11、mprove traditional inventory models with the aid of multi- layer BP neural network so as to acquire much more satisfactory optimum tactics of inventory.</p><p>　　2.THE LIMITATION TRADITIONAL INVENTORY OPTIMU

12、M MODEL</p><p>　　Before the strategic alliance relationship among the upstream and downstream enterprises comes into being, there is only a single material flow. The operational mechanics is shown below:<

13、/p><p>　　Under the operational mechanies of traditional supply chain(as show in figurel)，making inventory optimurn models; because of the lack of the necessary information, have to utilize probabilistic models

14、 to fit the changes of requirements based on the information of statistics. Now we give a simple single period random inventory model: </p><p>　　In this model:</p><p>　　E[T (y)] :The value of ex

15、pectation of the total cost of inventory; c :The manufacture(or purchase)cost of per product; h: The inventory cost of per product; p :The punishment cost for shorts of per product; x: The opening stock; y :The stock o

16、btained at opening; ξ: The demand during this epoch, it is a random variable; φ(ξ):The probability density function of ξ.</p><p>　　In order to minimize the value of E[T(y)],namely，this must have.Following

17、the method of derivation formulation which obtains parameter argument, we will get ;If give the value of c , h , p ,we can get the optimum value of stock y' ;also we can get the optimum inventory tactics during this

18、epoch.</p><p>　　As referred above, this traditional model is made under the insufficient information, so it is essential to lead many premise hypotheses, delimitate the application range, so this kind of mod

19、el is difficult to accord with the application in the real-world. The main problem focus on the probability density functions of ξ .From the analysis above we know the factors which affect the random variable ξ are a mul

20、ti-variable nonlinear relationship; such as: the price of product, the change of marketin</p><p>　　3.MODEL IMPROVEMENT BASED ON THE MECHANIC OF SUPPLY CHAIN MANAGEMENT</p><p>　　The direct and

21、 profound effect to the enterprise by the think of SCM is the change of decision mode: Change from the traditional, blocked longitudinal think mode into transversal, opening think mode. With the developing of IT and logi

22、stics technology, enterprise may have the ability to realize the translation based the Intranet\Extranet, Internet, and EDI technology .The operational mechanic of enterprise based on the SCM is shown below:</p>&

23、lt;p>　　According to the figure above, enterprises to one specific SC may share some important information which is the business secret for enterprise under the information enterprise traditional SC. With this inform

24、ation enterprise can improve prediction for inventory, marketing, the precision of etc.</p><p>　　3.1 MULTIPLE-LAYER BP NEURAL NETWORK</p><p>　　(1) Generalization of BPNN</p><p>　　Ar

25、tificial Neural Network, ANN for short, is an information-processing paradigm inspired by the way the densely interconnected, parallel structure of the mammalian brain processes information. In other words, artificial

26、neural networks are collections of mathematical models that emulate some of the observed properties of biological nervous systems and draw on the analogies of adaptive biological learning. The key element of the ANN pa

27、radigm is the novel structure of the information processin</p><p>　　There are multitudes of different types of ANNs, and BP Neural Network is a multi-layers back propagation neural network, which is trained

28、with the backpropagation of error algorithm. According to the simple structure and the considerable application, BPNN is the most popular ANN at the present.</p><p>　　(2)The basics of multi-layers BPNN</p

29、><p>　　BPNN is typically organized in layers. Layers are made up of a number of interconnected 'nodes' which contain an 'activation function'. Patterns are presented to the network via the '

30、input layer', which communicates to one or more 'hidden layers' where the actual processing is done via a system of weighted 'connections'. The hidden layers then link to an 'output layer'

31、 where the answer is output as shown in the graphic below:</p><p>　　The other two elements of BPNN are the propagation funtions of fi ,gi and the interconnection weights between the namely the weights :wij

32、, sij，and the value of threshold: θi,βi .The relationships of these elements are determinated by the equations as follow:</p><p>　　BPNN contains form of 'learning rule' which modifies weights of the

33、 connections according to the input patterns. Although there are many different kinds of learning rules, what BPNN uses the most often is the delta rule. With the delta rule, 'learning' is a supervised process th

34、at occurs with each cycle or 'epoch' through a forward activation flow of outputs, and the backwards error propagation of weight adjustments.</p><p>　　3.2 Making prediction BPNN model based on the ra

35、ndom variable ξ</p><p>　　As to the inventory optimum model, the key element is to fitting the change of the random variable of demand ξ, at the same time, the factors which affect the demand are variable, in

36、 this sense, it is also the most difficult process in the model. On the other hand enterprises which belonged to one specific SC can share some important information, due to the win-win relationships among these enter

37、prises. The information such as: the operational plan, the marketing intelligence etc. These facto</p><p>　　The key component for making inventory prediction BPNN is the choice of influence factors and the q

38、uantification of them. First of all the criteria for the choice of factors must lie on the contribution rate for ξ，then we will also take account of the feasibility of quantification. Now we will give a specific triple-l

39、ayers BPNN model based on the actual inventory condition of a steel corporation to predict the change of the demand of the steel plate.</p><p>　　This steel corporation is also a link in a supply chain, so it

40、 can get some specific information from its strategic cooperators. Now we select the demand of previous epoch: x1，the price of this epoch:x2，the internal rate of return of total steel vocation: x3，the factor of season c

41、hange: x 4,making the four factors as input layer; making the demand velocity v,the substitution of ξ，as output layer; At the same time, the number of the 'hidden layer' neuron should depend on the optimization

42、method</p><p>　　With some sampled data, we can select a suitable transfer function and train this model. In the process of training, we can use the ANN tools provided by MATLAB. Once the model is trained to

43、a satisfactory level, we can utilize it to predict the change of this corporation's inventory demand. To do this, we can get the next demand based on the current data.</p><p>　　4.CONCLUSION</p>&l

44、t;p>　　According to the anaaysis above, it is difficult to describe adequately the relationship of the factors which affect the demand of inventory with conventional approaches. Also we know that the ANNs are goo

45、d at solving problems that do not have an algorithmic solution or for which an algorithmic solution is too complex to be found. Summarily, The ANNs model, as to predicting the change of inventory demand, is a suitable a

46、pproach at the prrsent ,especially for BPNN model.</p><p><b>　　附錄二 中文譯文</b></p><p>　　供應鏈管理環(huán)境下的庫存優(yōu)化</p><p>　　云俊 鄢冰 趙玉瑋</p><p>　　武漢理工大學管理學院 郵編：430070<

47、/p><p>　　摘要：傳統(tǒng)供應鏈模式下的庫存優(yōu)化由于缺乏必要的信息，在模型的求解過程中難以得到符合實際要求的最優(yōu)解。本文分析了傳統(tǒng)企業(yè)庫存優(yōu)化與供應鏈管理環(huán)境下庫存優(yōu)化的運作機理，提出在供應鏈管理環(huán)境下可以借助多層BP神經(jīng)網(wǎng)絡改進傳統(tǒng)庫存模型，以得到更為滿意地最優(yōu)庫存策略。并依據(jù)某一鋼材現(xiàn)貨公司的庫存情況給出具體的應用。</p><p>　　關(guān)鍵詞：供應鏈 供應鏈管理 庫存 BP神經(jīng)網(wǎng)絡 優(yōu)

48、化</p><p><b>　　1.引言</b></p><p>　　供應鏈管理（簡稱SCM）是當今的一個熱門話題。這個詞來自關(guān)于作為一個特定的公司是如何組織聯(lián)系在一起的一幅圖片。供應鏈管理的想法是采用整體的方法來管理整個信息流，材料和來自于原材料供應商的服務通過工廠和倉庫直到最終的客戶。成功的供應鏈管理需要有一個一體化的系列活動納入一個緊密無間的過程。但是，在供應鏈

49、的每一個環(huán)節(jié)必然有一些延誤和一些不確定性，因此必須保持必要的庫存。相反，對企業(yè)來說存貨實際上是一種浪費。國內(nèi)外專家在庫存優(yōu)化領(lǐng)域已取得了很大的研究，做了許多庫存優(yōu)化的模型。所有這些模式在供應鏈管理的思想應運而生之前已經(jīng)取得了，但這些模型沒有考慮上游和下游企業(yè)。這些作為稀缺的信息優(yōu)化模型僅僅利用概率模型來適應信息統(tǒng)計基礎(chǔ)上需求的變化。通常情況下，通過這種方式制作的模型因為太過復雜而很難操作。另一方面，影響存貨清單的各因素之間的關(guān)系是非線性

50、的，因此很難作出一個定量和明確的數(shù)學關(guān)系，而且這些最佳的成果也不能滿足實際應用。</p><p>　　人工神經(jīng)網(wǎng)絡本身的自我學習和多映射的能力，可以探索復雜系統(tǒng)，使復雜的模型簡單化。在人工神經(jīng)網(wǎng)絡里，隱藏在網(wǎng)絡中的信息所作的聯(lián)系的神經(jīng)元，它可以處理多種定量關(guān)系。即神經(jīng)網(wǎng)絡是一個大規(guī)模并行計算模型，它的特點： </p><p>　　很大程度的魯棒性和容錯性；

51、 </p><p>　　隨時準備處理與一般非線性系統(tǒng)的相關(guān)問題；</p><p><b>

52、　　生物物理影響。</b></p><p>　　因此，對于非線性問題，人工神經(jīng)網(wǎng)絡是一個很好的分析工具。本文將提出在多層次的BP神經(jīng)網(wǎng)絡的幫助下，來改進傳統(tǒng)的庫存模型以獲得更令人滿意的優(yōu)化庫存。</p><p>　　2.傳統(tǒng)庫存優(yōu)化模型的局限性</p><p>　　上游和下游企業(yè)形成之前的戰(zhàn)略聯(lián)盟關(guān)系，只有一個單一的物質(zhì)流。運行機制如下圖所示：</

53、p><p>　　基于傳統(tǒng)供應鏈的運行機制(如圖所示) ，由于缺乏必要的信息，庫存決策優(yōu)化模型必須利用概率模型來適應信息統(tǒng)計基礎(chǔ)上需求的變化?，F(xiàn)在，我們給一個簡單的單周期隨機庫存模型：</p><p><b>　　在這個模型中：</b></p><p>　　E[T(y)]：價值期望的總費用清單； c：每種產(chǎn)品制造（或購買）的費用；h：每個產(chǎn)品的庫存成

54、本；p：缺少每個產(chǎn)品的懲罰成本；x:開放的股票；y：該股在開放時所得；ξ：在這個時期，它為隨機變量；φ（ξ）:概率密度函數(shù)。</p><p>　　為了盡量減少價值期望的總費用清單的價值，即：使價值期望的總費用清單最小，必須使。通過推導制定的方法得到參數(shù)的論點，我們將得到：；如果提供每種產(chǎn)品制造（或購買）的費用，每個產(chǎn)品的庫存成本，缺少每個產(chǎn)品的懲罰成本的價值，我們能獲得該股在開放時所得的最佳的價值股票，還可以得到

55、在這個時代的最佳的庫存策略。</p><p>　　正如上面提到的，這種傳統(tǒng)模式下取得的資料不足，涉及到許多相關(guān)的應用范圍，所以這是必不可少的前提假設，因此這種模式是難以符合實際應用的?，F(xiàn)在的主要問題集中在隨機變量的概率密度函數(shù)中。從上述分析我們知道影響隨機變量的因素是多變量非線性關(guān)系；如：產(chǎn)品的價格，銷售季節(jié)的變化，內(nèi)部收益率的總和。當然，對于一個特定的企業(yè)、影響因素可能是可變的。因此，在這個時期的隨機變量可能不

56、符合一個確定的概率分布，以及以這種方式獲得庫存的最優(yōu)戰(zhàn)略可能不符合現(xiàn)實的要求。</p><p>　　3.基于機械供應鏈管理上的模型改進</p><p>　　直接和深遠影響到企業(yè)的供應鏈變化的思考的決策模式: 改變傳統(tǒng)模式，阻止縱向思考模式進入橫向，縱向的思考模式打開。隨著IT和物流技術(shù)的發(fā)展，基于內(nèi)聯(lián)網(wǎng)\聯(lián)網(wǎng)，互聯(lián)網(wǎng)和電子數(shù)據(jù)交換技術(shù)，企業(yè)可能有能力實現(xiàn)翻譯。在供應鏈管理的基礎(chǔ)上機械業(yè)務的

57、企業(yè)如下所示：</p><p>　　根據(jù)上圖中，屬于一個具體供應鏈的企業(yè)可以分享一些重要的信息，這些信息在傳統(tǒng)供應鏈下是每個企業(yè)的商業(yè)秘密。有了這一信息的企業(yè)可以提高庫存的預測精度，銷售等。</p><p>　　3.1多層BP神經(jīng)網(wǎng)絡</p><p>　?。?）BP神經(jīng)網(wǎng)絡的摘要概括</p><p>　　人工神經(jīng)網(wǎng)絡，人工神經(jīng)網(wǎng)絡是一種信息處

58、理模式啟發(fā)，通過密集的相互聯(lián)系，哺乳動物大腦處理信息的平行結(jié)構(gòu)。換言之，人工神經(jīng)網(wǎng)絡的集合的數(shù)學模型，模擬的一些觀測特性的生物神經(jīng)系統(tǒng)，并利用類比的自適應生物學習。神經(jīng)網(wǎng)絡模式的關(guān)鍵因素是新型結(jié)構(gòu)的信息處理系統(tǒng)。它是由大量的高度聯(lián)結(jié)處理單元，類似于捆綁在一起，以加權(quán)聯(lián)系，類似于突觸。這一模式的優(yōu)勢尋找一個合適的預測模型庫存清單。</p><p>　　有眾多不同類型的人工神經(jīng)網(wǎng)絡和BP神經(jīng)網(wǎng)絡，這是進行了反向誤差算

59、法的訓練。根據(jù)簡單的結(jié)構(gòu)和大量的應用，人工神經(jīng)網(wǎng)絡是目前最流行的神經(jīng)網(wǎng)絡。</p><p>　?。?）基本的多層BP神經(jīng)網(wǎng)絡</p><p>　　通常BP神經(jīng)網(wǎng)絡的層次是有組織的。層是由若干包含一個'激活功能'的相互關(guān)聯(lián)的'節(jié)點'組成。模式通過“輸入層”提交給網(wǎng)絡，“輸入層”通過一個系統(tǒng)連接的加權(quán)對一個或更多的隱藏層進行實際加工。隱藏層然后鏈接到一個‘輸出層

60、’，在那里輸出所顯示的圖形如下：</p><p>　　BP神經(jīng)網(wǎng)絡的其他兩個要素是傳播fi，gi功能和神經(jīng)元之間的互連權(quán)重，即重：Wij，sij，和閾值的價值：θi，βi。這些元素之間的關(guān)系程度由方程式如下:</p><p>　　BP神經(jīng)網(wǎng)絡包含一些通過輸入模式來修改權(quán)的連接的某種形式的'學習規(guī)則'。雖然有許多不同類型的學習規(guī)則，但三角洲規(guī)則是BP神經(jīng)網(wǎng)絡用的最常見的學習規(guī)

61、則。在三角洲規(guī)則里， '學習'是出現(xiàn)在每個周期或'時代'的通過產(chǎn)出流動激活以及重量調(diào)整誤差，向后傳播的一個監(jiān)督的過程。</p><p>　　3.2在隨機變量的基礎(chǔ)上制作BP神經(jīng)網(wǎng)絡預測模型</p><p>　　作為庫存優(yōu)化模型、關(guān)鍵元素是適應隨機變量需求的變化。 同時，影響需求的因素是可變的，在這個意義上說，他也是模型中最艱難的過程。另一方面，因為這些企業(yè)雙

62、贏的關(guān)系，屬于一個具體供應鏈的企業(yè)可以分享一些重要的信息。如:操作計劃,營銷情報等信息。這些因素是非線性，為了使庫存優(yōu)化相當精確，我們可以利用三重層BP神經(jīng)網(wǎng)絡預測的變化著的預測模型。</p><p>　　制作BP神經(jīng)網(wǎng)絡庫存預測的關(guān)鍵部件是因素和量化的選擇。隨即變量首先要求選擇的因素必須符合在隨機變量的基礎(chǔ)上制作BP神經(jīng)網(wǎng)絡預測模型，然后，我們也將考慮到定量的可行性。現(xiàn)在我們根據(jù)鋼鐵企業(yè)實際的庫存條件給出一個具

63、體的三重層BP神經(jīng)網(wǎng)絡模型來預測鋼板的需求。</p><p>　　鋼鐵公司也是供應鏈的一個鏈接點，因此它可以從它的戰(zhàn)略合作者得到一些具體信息?，F(xiàn)在，我們選擇以往時代的需求：x1，這個時期的價格：x2，內(nèi)部收益率的鋼鐵行業(yè)的總數(shù)：x3，季節(jié)因素的變化：x4，四個因素的輸入層；決策速度的需求v，替代的隨機變量，作為輸出層；同時，隱層神經(jīng)元的數(shù)目應取決于我們所使用的優(yōu)化方法。該模型的結(jié)構(gòu)如下所示： </p>

64、<p>　　在一些采樣數(shù)據(jù)里，我們可以選擇一個合適的傳遞函數(shù)并且培訓這種模式。在這個培訓過程中，我們可以利用矩陣實驗室提供的神經(jīng)網(wǎng)絡工具。一旦模型訓練達到令人滿意的水平，我們可以利用它來預測本公司的庫存需求的變化。</p><p><b>　　4.結(jié)論</b></p><p>　　根據(jù)上述分析，很難適當描述傳統(tǒng)做法下影響庫存需求的因素之間的關(guān)系。另外我們