臺灣博碩士論文加值系統

『長鞭效應』（Bullwhip Effect）指在一個上、中、下游的體系中，下游些微的需求變動，可能造成中游的需求大幅度的變動，進而造成上游的需求變動更嚴重。亦即下游的需求變動是以放大倍數的方式向其上游傳遞，傳遞的層級越遠，造成的影響越大。
綜觀目前的企業生產情形，其產品型態大多已轉型為多種少量、生命週期與訂單交期日短、及採購與生產全球化的現象。供應鏈上游廠商為了因應終端產品的短生命週期及快速回應顧客，以致於造成各層廠商庫存的擠壓及需求放大，如此便使得製造過多的成品零件及物料的缺貨損失。因此本文針對供應鏈需求預測問題，藉由啤酒遊戲軟體，設定供應鏈中存貨政策、外部顧客需求、前置時間三個供應鏈的參數值，以模擬出二十四種供應鏈情境，並在模擬的供應鏈各種情境中，收集各階層訂單標準差。本研究分為兩種模型進行資料分析，第一種模型是將所有的解釋變數(存貨政策、外部顧客需求、前置時間)，以工廠和零售商訂單標準差的距離作為解釋變數，分別利用倒傳遞類神經（Back Propagation Network, BPN）、多元適應性迴歸（Multivariate Adaptive Regression Splines, MARS）兩種單一分析工具，來驗證此模式；第二種模型則是先透過MARS粹取出重要的解釋變數，再將重要的解釋變數分別以MARS與BPN進行資料分析，以驗證此模式。實驗結果，第一種模型及第二種模型中，BPN方法有較好的表現，此外亦可以得知兩階段的模型比單一階段的預測表現還好。此結論希望可以針對長鞭效應的預測方法作評估，降低長鞭效應所帶來的影響，提供企業供應鏈管理之用。

The essence of Bullwhip Effect is the reciprocation among upper, middle, and lower sections in an industrial system. Any slight variation of need in lower section could make a substantial alteration in middle section; proceed to higher section a gigantic needing shift. In other words, the needing shift of lower section magnifies the transference to upper section, higher the transference reached, bigger the influence impacted.
Making a comprehensive survey of industrial yield, the types of products are transformed into various, low quantity, shortened product life-cycle and delivery time, and globalization in both purchasing and manufacturing. In supply chain, the-upper section-manufactory in behalf of dealing with the short terminal product life-cycle and fast-responding customer service, so that to cause every sections magnifying the manufactory’s stocks extrusion and demand, thereupon making excess end items and materials to run out of stock. Furthermore, this essay focuses on issue of predicting supply chain need. By the way of applying the beer game software to set the three parameters in supply chain: stock policy, external customer needs and lead time. In order to stimulate twenty-four types of supply chain circumstances, and collecting the value of all sections in the stimulated supply chain.
This research analyzed the data on using two models. The first model is taking all the explanatory variables, including stock policy, external customer needs, and lead time, into consider. This uses the range of order standard deviation between factory and retailer as explanatory variable. Utilize two unitary analytical tools, Back Propagation Network and Multivariate Adaptive Regression Splines, separately, in order to prove . The outcome is BPN has better predict performancethan MARS. The second model extracts the essential explanatory variables through MARS in advance. Then, data analyze them with MARS and BPN, respectively, so as to prove . Then this research found out that two steps predict model are better than just one step predict model. This research report is trying to estimate the predicting method of Bullwhip Effect, to use as diminishing the impact it brings out. This essay provides enterprise for managing supply chain.

目錄
中文摘要 II
ABSTRACT III
目錄 VI
表目錄 VIII
圖目錄 IX
第一章 緒論 1
1.1研究背景與研究動機 1
1.2研究目的 2
1.3研究範圍 3
1.4研究流程 4
第二章 文獻探討 5
2.1 供應鏈管理 5
2.1.1 供應鏈管理的起源 5
2.1.2 供應鏈管理的定義 6
2.2 長鞭效應 8
2.2.1長鞭效應的成因 8
2.2.2長鞭效應的影響 11
2.2.3長鞭效應的因應策略 12
2.2.4長鞭效應指標 18
2.3預測方法 19
2.3.1 預測概論 19
2.3.2預測的定義 19
2.3.3預測的方法 21
2.3.4預測績效評定方法 29
2.3.5類神經的網路模式 29
2.3.6多元適應性雲形迴歸的介紹 31
2.4啤酒遊戲介紹 33
2.4.1傳統啤酒遊戲之缺失 33
2.4.2電腦化啤酒遊戲 34
第三章 研究方法與內容 37
3.1啤酒遊戲模擬 37
3.1.1問題描述 37
3.1.2模式假設 37
3.1.3啤酒遊戲模式情境選擇 39
3.1.3 預測流程 44
3.2倒傳遞類神經網路 45
3.3多元適應性雲形迴歸 47
第四章 驗證結果分析 50
4.1單一分析工具驗證分析 50
4.1.1 BPN模式驗證分析 50
4.1.2 MARS模式驗證分析 52
4.2 二階段模型驗證分析 53
4.2.1MARS模式驗證分析 53
4.2.2 BPN模式驗證分析 54
4.3 變異數分析 57
第五章 結論與建議 60
5.1研究結論 60
5.2研究限制 60
5.3後續研究建議 61
參考文獻 62
表目錄
表2.1 供應鏈管理的定義 8
表2.2 長鞭效應造成因素 11
表2.3 長鞭效應的影響 12
表2.4 長鞭效應因應對策相關研究整理 13
表2.5 其他長鞭效應的相關文獻整理 15
表2.6 長鞭效應衡量指標相關文獻 18
表2.7 預測方法比較 24
表2.7 預測方法比較[續] 25
表2.8 類神經網路預測方法其他預測方法比較相關文獻 26
表2.9 MARS與其他預測方法比較相關文獻 28
表2.10 服務水準及M值 35
表3.1 供應鏈情境設定 40
表3.2 供應鏈情境 41
表3.3 變數編碼 42
表4.1 BPN模式於不同參數組合之診斷結果 51
表4.3 MARS模式篩選之顯著影響變數 53
表4.4 MARS迴歸統計值檢定結果 54
表4.5 BPN模式於不同參數組合之診斷結果 55
表4.6 兩種模式分析結果 56
表4.7 變異數分析結果 57
表4.8 迴歸分析結果 58
圖目錄
圖1.1 研究流程 4
圖3.1 啤酒遊戲供應鏈設計 38
圖3.2 啤酒遊戲流程圖 43
圖3.3 預測流程 44
圖3.4 類神經網路架構圖 45
圖3.5 單變數BF折線圖 48
圖4.1 {3-8-1}之BPN訓練模式樣本RMSE趨勢圖 52
圖4.2 {2-4-1}之BPN訓練模式樣本RMSE趨勢圖 56

參考文獻
[1]王立志，“系統化運籌與供應鏈管理”，滄海書局，1999年初版。
[2]李瓊英，長鞭效應下最佳存貨策略的決定與環境變數效果的探討，國立台灣大學商學研究所碩士論文，1999年。
[3]春日井博，需求預測入門，方世榮 (校閱)，書泉書局，1988年。
[4]徐桂祥，「灰色系統在商情預測上之研究」，雲林技術學院資訊管理技術研究所碩士論文，1997年。
[5]徐壽政，長鞭效應之情境分析模式，國立交通大學工業工程與管理研究所碩士論文，1998年。
[6]葉怡成，類神經網路模式應用與實作，儒林出版社，1993年。
[7]許乃文，「灰色理論及類神經網路應用於雲林地區地層下陷之研究」，國立成功大學土木工程學系碩士論文，2000年。
[8]許峻源，「類神經網路與MARS於資料探勘分類模式之應用」，輔仁大學應用統計研究所碩士論文，2001年。
[9]董興國，「我國半導體IC產業之系統動態學模式研究」，私立元智大學管理研究所碩士論文，1997年。
[10]曾中文，拉式供應鏈(s, Q)存貨政策管理之探討，逢甲大學工業工程學研究所碩士論文，2000年。
[11]趙嬙，灰色預測理論應用於汽車產業預測之研究－以台灣、大陸市場為例，朝陽科技大學企業管理研究所碩士論文，2003年。
[12]鄭穎聰，供應鏈長鞭效應因應政策之研究，國立台北科技大學生產管理研究所碩士論文，2000年。
[13]盧儒瓊，預測週期和預測方式對長鞭效應的影響，國立交通大學工業工程與管理研究所碩士論文，2001年。
[14]Andrew S. Caplin, “The variability of aggregate demand with (S, s) inventory policies,” Econometrica, Vol.53, No. 6, pp. 347-356, 1985.
[15]Beamon, B. M., “Supply Chain Design & Analysis: Models & Methods”, International Journal of Production Economics, Vol.23, No. 55, pp. 281-294,1998.
[16]Bechtel, C., and Jayaram, J., “Supply Chain Management: A Strategic Perspective,” The International Journal of Logistics Management, Vol. 8, No. 1, pp.15-34 , 1997.
[17]Blackburn,J.D., The quick response movement in the apparel industry : a case study in time-compressing supply chains, in time-based competition : The next Battleground in American Manufacturing, Irwin, Homewood, IL, pp. 133-156, 1991
[18]Blackburn and R.Millen, ”Improved Heuristics for Multi Stage Requirements Planning Systems”, Management Science, Vol.28, No.1, pp.44-58, 1982.
[19] Chu, C. H. and Widjaja, D., “Neural Network System for Forecasting Method Selection,” Decision Support System, Vol.12, Iss.1 , pp. 13-24,1994.
[20]Cooper, M. C. and Ellram, L. M., “Characteristics of Supply Chain Management and the Implications for Purchasing and Logistics Strategy”, The International Journal of Logistics Management, Vol. 4, No. 2, pp. 13-24,1993.
[21]Cybenko, G., “Approximation by Superpositions of a Sigmoidal Functions,” Mathematics of Control, Signals, and Systems,Vol.2, No. 3, pp.303-314, 1989.
[22]De Gooijer, J. G., Ray, B. K., and Krager, H., “Forecasting Exchange Rates Using TSMARS,” Journal of International Money and Finance, Vol.17, Issue 3, pp.513-534, 1998.
[23]Donlebell and O.J. Krasner,“Selecting Environmental Forecasting from Business Planning Requirements ”Academy of Management Review, Vol.19, No. 5, pp.373-383 ,1977.
[24]Ellram, L. M., “The Use of the Case Study Method in Logistics Research”, Journal of Business Logistics, Vol. 17, Iss. 2, pp. 93-138, 1996.
[25]Forrester,Jay W.“Market Growth as Influenced by Capital Investment” Industrial Management Review, Vol. 6, No. 2, pp.345-368, 1968.
[26] Frank Chen, Zvi Drezner, Jennifer K. Ryan, David Simchi-Levi, “Quantifying the Bullwhip Effect in a Simple Supply Chain: The Impact of Forecasting, Lead Times, and Information,” Management Science, Vol. 46, No. 3, pp.233-242, 2000.
[27]Friedman, J. H., “Multivariate Adaptive Regression Splines (with discussion), ” Annals of Statistics, Vol.19, No. 1, pp. 1-141 ,1991.
[28]Friedman, J. H. and Roosen, C. B., “An Introduction to Multivariate Adaptive Regression Splines,” Statistical Methods in Medical Research, Vol.4, No.3, pp.197-217, 1995.
[29]Handfield, R. B., and Nichols, E. L., “Introduction to Supply Chain Management ” ,Prentice-Hall Internation Editions, pp. 345-367, 1999.
[30]Hecht-Nielsen, R., “Neurocomputing”, Menlo Park Editions, pp. 212-240, 1990.
[31]Herbig, P., J. Milewicz and J. E. Golden , "Forecasting: Who, What, When, and How", Journal of Business Forecasting, Vol.12, No.2, pp. 16-21, 1993.
[32]Hornik, K., Stinchcombe, M. and White, H, “Multilayer feedforward networks are universal approximators,” Neural Networks, Vol. 2, Iss. 5 , pp.359-366, 1989.
[33]Hopfield, J. and Tank, D., “Neural Computations of Decisions in Optimization Problems,” Biological Cybernetics, Vol. 51, No.10, pp.141-152,1985.
[34]Houlihan, J. B., “International Supply Chain Management”, International Journal of Physical Distribution and Materials Management, Vol. 15, Iss. 1, pp. 22-38, 1985.
[35]Huan, J. S. and T. L. James,and P. W. Trefor,“Using Neural Networks to Predict Component Inspection Requirement for Aging Aircraft”, Computers Ind, Engng, Vol. 30, No. 2, pp.257-267, 1996.
[36]Johnson, J. C., and Wood, D. F., “Contemporary Logistics”, Prentice-Hall Internation Editions, 6th Ed, pp.124-150, 1996.
[37]Kahn, J., Inventories and the Volatility of Production. The Amer. Econom. Rev., Vol. 77,No. 4, pp. 667-679,1987.
[38]Kalakota, R. and Whinston, A. B., Frontiers of Electronic Commerce, Addison-Wesley Publishing, pp. 67-89, 1996.
[39]Kelle, P., and Milne, A., “The Effect of (s,S) Ordering Policy on the Supply Chain”, Int. J. Production Economics, Vol.59, Iss:1-3, pp. 113-122,1999.
[40]Kuhnert, P. M., Do, Kim-Anh, and McClure, R., “Combining non-parametric models with logistic regression: an application to motor vehicle injury data,” Computational Statistics and Data Analysis, Vol.34, Iss. 3, pp.371-386, 2000.
[41]LaLonde, Bernard, J. and Masters, James M., “Emerging Logistics Strategies: Blueprints for the Next Century”, International Journal of Physical Distribution and Logistics Management, Vol.24, No.7, pp.35-47, 1994.
[42]Lambert, D. M., and Cooper, M. C., “Issues in Supply Chain Management”, Industrial Marketing Management, Vol. 29, Iss. 1, pp. 65-83, 2000.
[43]Larry, M., Efraim, T., and Robert R. T., Neural Network Fundamentals for Financial Analysts. In Robert R. Trippi, and Efraim Turban（ED.）, Neural Networks: in Finance and Investing, Chicago, Probus Publishing Company, pp.3-25, 1993.
[44]Lee, H. L., Padmanabhan, V., Whang, S., “Information Distortion in a Supply Chain：The Bullwhip Effect”, Management Science, Vol. 43, Iss. 4, pp. 546-558, 1997.
[45]Lewis, P. A. W. and Stevens, J. G., “Nonlinear Modeling of Time Series Using Multivariate Adaptive Regression Splines（MARS）,” Journal of American Statistical Association, Vol.86, No.416, pp. 864-877, 1991.
[46]Lippmann, R. P., “An Introduction to Computing with Neural Networks,” IEEE ASSP Magazine,Vol.56, No.22, pp. 4-22, 1987.
[47]Marshall, G., Grover, F. L., and Henderson, W. G., Hammermeister, K. E., “Assessment of Predictive Models for Binary Outcomes: An Empirical Approach Using Operative Death from Cardiac Surgery,” Statistics in Medicine,Vol.13, Iss 15, pp.1501-1511,1994.
[48]Metters, R., “Quantifying the bullwhip effect in supply chains,” Journal of Operations Management, Vol 15, No. 2, pp. 89-100, 1997.
[49]Model,J.M.,”A Preude to Neural Network : Adaptive and Learning System”, Prentice-Hall, pp. 23-56, 1994.
[50]Moshiri, S. and N. Cameron, “Neural Network versus Econometric Models in Forecasting Inflation”, Journal of Forecasting, Vol.19, No.3, pp.201-217, 2000.
[51]Nguyen-Cong V., Van D. G. and Rode, B. M., “Using Multivariate Adaptive Regression Splines to QSAR Studies of Dihydroartemisinin Derivatives,” Eur. J Med. Chem., Vol.31, Iss 10, pp. 797-803, 1996.
[52]Nguyen, N. and A. Cripps, “Predicting Housing Value：A Comparison of Multiple Regression Analysis and Artificial Neural Networks”, Journal of Real Estate Research, Vol.22, No.3, pp. 313-336, 2001.
[53]Patrovic, D., R. Roy, and R. Petrovic, “Modeling and Simulation of a Supply Chain in an Uncertain Environment,” European Journal of Operational Research, Vol. 109, Iss. 2, pp.299-309, 1998.
[54]Remer,D.S. and Jorgens, C., “Ethylene Economics and Production Forecasting in a Changing Environment ”, Engineering & Process Economics, Vol.3 lss.4, P267-278, 1978
[55]Ross, D. F., “Competing Through Supply Chain Management：Creating Market—Winning Strategies through Supply Chain Partnerships”, Chapman & Hall, pp.239-259, 1998.
[56]Simchi-Levi, D., Kaminsky, P. and Simchi-Levi, E., Designing And Managing The Supply Chain, and Case Studies, Boston, McGraw-Hill International Edition., pp. 198-210, 2000.
[57]Shaaf, M. “Predicting Recession Using the Yield Curve：An Artificial Intelligence and Econometric Comparison”, Eastern Economic Journal, Vol.26, No.2, pp. 171-190, 2000.
[58]Sterman, J. D., Modeling Managerial Behavior: Misperceptions of Feedback in a Dynamic Decision Making Framework. Management Science, Vol. 35, No. 3, pp. 321-339,1989.
[59]Stone, G., Chan, D., Kuhnert, P. M., and Cameron, M., “Some Experience in the Analysis of Large and Complex Datasets,” Computing Science and Statistics,
Proceedings of the Second World Congress of the IASC, Vol.29, No.2, pp. 134-1401997.
[60]Sunil Chopra, and Peter Meindl, “ Supply Chain Management :Strategy,Planning,and Operation”,Prentice Hall,Inc., pp. 234-250, 2001
[61]Tang, Z., Almeida, C. and Fishwick, P. A., “Time Series for Using Neural Networks vs Box-Jenkins Methodology,”Simulation, Vol.57, No.5, pp. 303-310, 1991.
[62]Tang, Z. and Fishwick, P. A., “ Feedforward Neural Nets as Models for Time Series Forecasting,” ORSA Journal on computing, Vol.5, No.4, pp.374-385, 1993.
[63]Thomas, D. J., and Griffin, P. M., “Coordinated Supply Chain Management ”,European Journal of Operational Research, Vol. 94, Iss. 1, pp. 1-15, 1996.
[64]Towill, D. R.,”Time compression and supply chain management-a guided tour”, Supply Chain Management, Vol.1, No.1, pp. 15-27, 1997.
[65]W. L., Fisher, N. I., Friedman, J. H., and Ryan, C. G., ”Statistical Techniques for the Classification of Chromites in Diamond Exploration Samples”. Journal of Geochemical Exploration, Vol.59, Iss.3, pp. 233-249,1997.
[66]Wong, F. S., “Time Series Forecasting Using Backpropagation Neural Networks,” Neurocomputing, Vol.2, Iss. 4, pp. 147-159, 1991.
[67]Zhang, G., Patuwo, B. E., and Hu, M. Y., “Forecasting with artificial neural networks: The state of the art,” International Journal of Forecasting, Vol.14, Iss. 1, pp.78-85, 1998.