發(fā)布時(shí)間：2018-05-10 18:41

  本文選題：電子商務(wù) + 閉環(huán)供應(yīng)鏈�。� 參考：《沈陽理工大學(xué)》2017年碩士論文

【摘要】：電子商務(wù)是科技信息發(fā)展的產(chǎn)物,它是以信息網(wǎng)絡(luò)技術(shù)為手段,以商品交換為中心的商務(wù)活動(dòng),是傳統(tǒng)商業(yè)活動(dòng)各環(huán)節(jié)的電子化、網(wǎng)絡(luò)化、信息化。牛鞭效應(yīng)是供應(yīng)鏈中需求信息被放大的現(xiàn)象,它使得供應(yīng)鏈系統(tǒng)的經(jīng)濟(jì)效益受損,嚴(yán)重時(shí)會(huì)導(dǎo)致整個(gè)供應(yīng)鏈的崩潰。電子商務(wù)的出現(xiàn)使得閉環(huán)供應(yīng)鏈網(wǎng)絡(luò)變得更加復(fù)雜,牛鞭效應(yīng)對(duì)其產(chǎn)生的影響也因此變得更加顯著。預(yù)測(cè)控制算法已經(jīng)在各個(gè)領(lǐng)域得到了廣泛的應(yīng)用,面對(duì)供應(yīng)鏈管理面臨的新的挑戰(zhàn),專家學(xué)者將預(yù)測(cè)控制算法引入到供應(yīng)鏈系統(tǒng)的管理中,并得到滿意的控制效果。動(dòng)態(tài)矩陣控制算法是預(yù)測(cè)控制算法中典型算法之一,該算法具有實(shí)現(xiàn)簡(jiǎn)單、控制靈活和應(yīng)用范圍廣等優(yōu)點(diǎn)。針對(duì)電子商務(wù)環(huán)境下供應(yīng)鏈管理中存在的問題,本文對(duì)動(dòng)態(tài)矩陣控制算法在閉環(huán)供應(yīng)鏈管理中的應(yīng)用進(jìn)行研究。牛鞭效應(yīng)是衡量供應(yīng)鏈系統(tǒng)最重要的性能指標(biāo),本文針對(duì)供應(yīng)鏈中存在的牛鞭效應(yīng)進(jìn)行深入研究。牛鞭效應(yīng)造成庫(kù)存成本增加、生產(chǎn)過量、客戶滿意度降低,它產(chǎn)生的主要原因有:需求預(yù)測(cè)不準(zhǔn)、價(jià)格波動(dòng)和訂單增加。為了直觀分析牛鞭效應(yīng),本文采用統(tǒng)計(jì)分析的方法對(duì)其實(shí)現(xiàn)量化并通過加強(qiáng)信息共享、減小波動(dòng)性、建立戰(zhàn)略伙伴關(guān)系來削弱和抑制牛鞭效應(yīng)帶來的負(fù)面影響和危害。動(dòng)態(tài)矩陣控制算法(DMC)包括預(yù)測(cè)模型、滾動(dòng)優(yōu)化和反饋校正三部分,文中描述了算法的實(shí)現(xiàn)步驟,針對(duì)系統(tǒng)參數(shù)的取值給出了選取原則和整定步驟,并使用MATLAB完成了算法編程實(shí)現(xiàn),然后分別在單輸入單輸出對(duì)象和多輸入多輸出對(duì)象的情況下進(jìn)行仿真實(shí)驗(yàn),實(shí)驗(yàn)結(jié)果表明系統(tǒng)參數(shù)的選取和整定步驟的有效性。將動(dòng)態(tài)矩陣控制算法應(yīng)用到電子商務(wù)環(huán)境下閉環(huán)供應(yīng)鏈管理中的實(shí)質(zhì)是通過算法的滾動(dòng)優(yōu)化和反饋校正,不斷調(diào)整供應(yīng)鏈中節(jié)點(diǎn)企業(yè)的訂貨量和生產(chǎn)量,最終使得供應(yīng)鏈系統(tǒng)中的牛鞭效應(yīng)得到削減和抑制。本文建立了雙渠道電子商務(wù)閉環(huán)供應(yīng)鏈網(wǎng)絡(luò)的動(dòng)態(tài)模型,利用卡爾曼濾波器對(duì)系統(tǒng)的狀態(tài)、不可測(cè)擾動(dòng)和噪聲進(jìn)行估計(jì),并推導(dǎo)出預(yù)測(cè)模型,確定系統(tǒng)目標(biāo)函數(shù),并提出了二次優(yōu)化的約束優(yōu)化算法,將供應(yīng)鏈動(dòng)態(tài)網(wǎng)路管理問題轉(zhuǎn)換成標(biāo)準(zhǔn)二次優(yōu)化問題進(jìn)行求解。通過仿真實(shí)驗(yàn)結(jié)果表明,將DMC算法應(yīng)用在電子商務(wù)環(huán)境下閉環(huán)供應(yīng)鏈的管理中,能為供應(yīng)鏈系統(tǒng)提供優(yōu)化的生產(chǎn)庫(kù)存策,增強(qiáng)了系統(tǒng)的穩(wěn)定性,提高了鏈中企業(yè)的競(jìng)爭(zhēng)力,最終實(shí)現(xiàn)了整體利益的最大化。
[Abstract]:E-commerce is the product of the development of scientific and technological information. It is a commercial activity with information network technology as the means and commodity exchange as the center. It is the electronic, networked and information of each link of the traditional commercial activities. Bullwhip effect is the phenomenon that the demand information is enlarged in the supply chain, which makes the economic benefit of the supply chain system damaged, and will lead to the collapse of the whole supply chain seriously. The emergence of electronic commerce makes the closed-loop supply chain network more complex, and the bullwhip effect becomes more significant. Predictive control algorithm has been widely used in various fields. In the face of new challenges faced by supply chain management, experts and scholars introduce predictive control algorithm into supply chain management, and obtain satisfactory control effect. Dynamic matrix control algorithm is one of the typical predictive control algorithms, which has the advantages of simple implementation, flexible control and wide application. In this paper, the application of dynamic matrix control algorithm in closed-loop supply chain management is studied. Bullwhip effect is the most important performance index to measure the supply chain system. Bullwhip effect leads to the increase of inventory cost, excessive production and decrease of customer satisfaction. The main reasons are: inaccurate demand forecast, fluctuating price and increasing order. In order to analyze the bullwhip effect intuitively, this paper uses the statistical analysis method to quantify the bullwhip effect and to weaken and restrain the negative influence and harm brought by the bullwhip effect by strengthening the information sharing, reducing the volatility and establishing the strategic partnership. The dynamic matrix control algorithm (DMC) consists of three parts: prediction model, rolling optimization and feedback correction. The implementation steps of the algorithm are described, and the selection principles and tuning steps are given for the selection of system parameters. The algorithm is programmed with MATLAB, and the simulation experiments are carried out in the case of single input, single output object and multiple input and multiple output object, respectively. The experimental results show that the system parameter selection and tuning steps are effective. The essence of applying the dynamic matrix control algorithm to the closed loop supply chain management in the electronic commerce environment is to adjust the order quantity and production capacity of the node enterprises in the supply chain through rolling optimization and feedback correction of the algorithm. Finally, the bullwhip effect in the supply chain system is reduced and restrained. In this paper, the dynamic model of the closed loop supply chain network of dual channel electronic commerce is established. The state, unmeasurable disturbance and noise of the system are estimated by Kalman filter, and the prediction model is derived, and the objective function of the system is determined. A constrained optimization algorithm for quadratic optimization is proposed to transform the dynamic network management problem of supply chain into a standard quadratic optimization problem. The simulation results show that the application of DMC algorithm in the management of closed loop supply chain under electronic commerce environment can provide the supply chain system with optimized production inventory policy, enhance the stability of the system, and improve the competitiveness of enterprises in the chain. Finally realized the overall benefit maximization.
【學(xué)位授予單位】：沈陽理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：F274;F724.6;TP13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李世卿;丁寶蒼;;基于動(dòng)態(tài)矩陣控制的雙層結(jié)構(gòu)預(yù)測(cè)控制的整體解決方案[J];自動(dòng)化學(xué)報(bào);2015年11期

2 孫建平;齊園園;;動(dòng)態(tài)矩陣控制在電加熱爐溫度控制中的應(yīng)用[J];計(jì)算機(jī)仿真;2013年06期

3 席裕庚;李德偉;林姝;;模型預(yù)測(cè)控制——現(xiàn)狀與挑戰(zhàn)[J];自動(dòng)化學(xué)報(bào);2013年03期

4 鐘恒;;供應(yīng)鏈牛鞭效應(yīng)的成因及其弱化策略的分析[J];經(jīng)營(yíng)管理者;2010年13期

5 吳響;張敏;馬震;;基于動(dòng)態(tài)矩陣控制算法的網(wǎng)絡(luò)控制系統(tǒng)的研究[J];中國(guó)新通信;2010年03期

6 鐘昌寶;聶茂林;徐永其;;基于灰色馬爾柯夫改進(jìn)模型預(yù)測(cè)供應(yīng)鏈獨(dú)立需求[J];情報(bào)雜志;2009年07期

7 董海;王宛山;李彥平;鞏亞東;;分布式MPC在網(wǎng)絡(luò)化制造環(huán)境下SCM中的應(yīng)用[J];系統(tǒng)仿真學(xué)報(bào);2007年06期

8 黃小原,盧震;多分銷中心供應(yīng)鏈模型及其牛鞭效應(yīng)的H_∞控制[J];中國(guó)管理科學(xué);2003年01期

9 葉丹,戰(zhàn)德臣,徐曉飛,李全龍,李風(fēng)華;多視圖集成化的動(dòng)態(tài)聯(lián)盟建模方法[J];高技術(shù)通訊;2000年02期

10 李少遠(yuǎn),席裕庚;多模型預(yù)測(cè)控制的平滑切換[J];上海交通大學(xué)學(xué)報(bào);1999年11期

，

本文編號(hào)：1870429