簡易檢索 / 詳目顯示

回結果列表
研究生： 陳泯璋
Ming-Chang Chen
論文名稱： 以類神經網路結合多代理人系統運用在生產規劃與存貨監控之個案研究
A Case Study on Applying Neural Networks and Multi-Agent Systems to Production Planning and Inventory Monitoring
指導教授： 蘇豐文
Von-Wun Soo
口試委員:
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 資訊系統與應用研究所
Institute of Information Systems and Applications
論文出版年： 2008
畢業學年度： 96
語文別： 中文
論文頁數： 56
中文關鍵詞： 多代理人 、類神經網路 、生產策略
外文關鍵詞： multi-agent, neural network, production strategy
相關次數： 點閱：1  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 在現今的商業環境競爭愈來愈激烈，企業的既定生產策略常常受到不確定因
    素的影響而需要作修正，例如銷售量不如預期、消費者需求改變等因素。因此如
    何快速因應環境的變化而修正既定的生產策略便突顯的相當重要。
    本研究在探討如何運用多代理人在生產規劃與庫存管控並以造紙業為例。傳
    統的造紙業生產模式由於各部門之間的資訊並沒有彼此傳遞，其生產量仍以過去
    的經驗判斷為主，各部門之間缺乏一套整合的工具導致生產變異居高不下。為了
    改善此問題，本研究運用代理人傳遞與溝通訊息的功能並輔以類神經網路預測，
    使各部門之間的資訊系統能達到整合，進而在生產策略能做更有效率的決策。
    本研究首先會進行系統分析獲得需求，在設計階段會開始規劃代理人的類別與功
    能。最後在實驗階段會以從案例公司獲得的實際資料，套入代理人程式當中並輔
    以類神經預測軟體進行實驗與比較分析結果。

    It is more and more competitive in modern commercial environment; the enterprises often adjust their production plans because of uncertain factors. For example, the sales fall short of their expectations, the demand of consumers change, etc. Therefore, it is very important to the enterprises to rapidly adjust their production strategy caused by the changes of extrinsic environment.
    This research treats of how to employ the agents on production planning and inventory monitor, and takes the paper manufacturing industry as an example. There is little cooperation among each department in the traditional paper manufacturing industry. Therefore, the production plans still depend on historical experiences. There is high variance of production because of its lack of integrated tools among each department. In order to improve this problem, the agent-based system which can integrate information is developed by neural network in this research. It can integrate information among each department and support more efficient decision-making.
    In this research, the system analysis is processed to get requirements at first, and capture the collaboration requirements among the various functions of planning, the categories and functions of agents are created in the designed stage. Finally, in the executive stage, the real data got from case enterprise are used to make an experiment.

    目錄 中文摘要...........................................................................................2 英文摘要...........................................................................................3 第一章 序論....................................................................................10 1.1. 研究動機與目的..................................................................10 1.2. 研究方法.............................................................................11 1.3. 研究範圍.............................................................................11 1.4. 章節闡述.............................................................................12 第二章 文獻探討.............................................................................14 2.1. 造紙業的生產模式.................................................................14 2.2. 代理人的功能與性質...........................................................14 2.3. 預測的相關方法..................................................................16 2.4. 類神經網路(Artificial Neural Network) .................................18 2.4.1. 類神經網路的要素............................................................18 2.4.1.1 處理單元(Processing Element ，PE) ......................18 2.4.1.2 層(Layer) ...............................................................19 2.4.1.3 網路(Network) .......................................................19 第三章 研究方法...........................................................................20 3.1. 系統分析階段(System Analysis Phase) .................................20 3.2. 系統設計階段(System Design Phase) ...................................21 3.2.1. 獲得目標.....................................................................22 3.2.2. 實施使用案例..............................................................23 3.2.3. 精煉系統內的角色.......................................................24 3.2.4. 建立代理人類別..........................................................25 3.2.5. 建立對話與組合代理人類別........................................26 3.2.6. 系統設計.......................................................................26 3.3. 系統實作階段(System Execution Phase) ...............................27 第四章 造紙業生產之模型分析與系統建立...................................28 4.1. 分析階段............................................................................28 4.2. 設計階段............................................................................30 4.2.1. 確立目標.....................................................................30 4.2.2. 實施使用案例..............................................................31 4.2.3. 精煉系統內的角色.......................................................33 4.2.4. 建立代理人類別..........................................................34 4.3. 實作階段............................................................................35 第五章 系統實作與實驗分析.........................................................37 5.1. 資料來源簡介.....................................................................37 5.2. 類神經網路的建置..............................................................37 5.2.1. 期初銷售預測的類神經模型........................................39 5.2.2. 期中銷售預測的類神經模型........................................39 5.2.3. 安全庫存的類神經模型...............................................39 5.2.4. 期末存貨的類神經模型...............................................40 5.3. 生產、銷售、庫存代理人間的運作與協調...........................40 5.4 數學公式...............................................................41 5.4.1. 銷售預測部份..............................................................41 5.4.2. 庫存管控部份..............................................................42 5.4.3. 生產規劃部份..............................................................42 5.5. 實驗過程與結果分析............................................................43 5.5.1. 實驗過程.......................................................................43 5.5.2. 實驗結果分析................................................................48 第六章 結論與未來展望.................................................................52 參考文獻.........................................................................................53 圖目錄 圖2.1. 計劃型生產流程........................................................................14 圖3.1. 加值鏈....................................................................................20 圖3.2. 事件流程圖..............................................................................21 圖3.3. 多代理人系統分析方法程序圖.....................................................22 圖3.4. 目標階層圖..............................................................................23 圖3.5. 使用案例圖..............................................................................23 圖3.6. 循序圖....................................................................................24 圖3.7. 角色模型.................................................................................25 圖3.8. 代理人類別圖...........................................................................26 圖3.9. 實作概念架構圖........................................................................27 圖4.1. 造紙業供應鏈...........................................................................28 圖4.2. 造紙業加值鏈圖........................................................................29 圖4.3. 造紙業生產流程圖....................................................................30 圖4.4. 生產目標階層圖........................................................................31 圖4.5. 使用案例圖..............................................................................32 圖4.6. 造紙業生產循序圖....................................................................33 圖4.7. 角色模型.................................................................................34 圖4.8. 代理人類別圖...........................................................................35 圖4.9. 實作架構圖..............................................................................36 圖5.1. 執行期初預測...........................................................................43 圖5.2. 執行期中預測...........................................................................44 圖5.3. 期中銷售預測結果....................................................................44 圖5.4. 2004~2005年期中預測結果.........................................................45 圖5.5. 2004~2005年安全庫存結果..........................................................46 圖5.6. 2004~2005年本研究生產量..........................................................46 圖5.7. 2004~2005年本研究期末庫存量...................................................47 圖5.8. 執行結果.................................................................................48 圖5.9 2004年生產量比較...................................................................50 圖5.10 2005年生產量比較...................................................................50 表目錄 表5.1. 成本係數表..............................................................................48 表5.2. 成本分析表..............................................................................51

    [1] Bradshaw, J. M., (ed.,)., “Software Agents”, Menlo Park, CA:AAAI/MIT Press, 1997.
    [2] DeLoach, S. A., “Analysis and Design Using MaSE and AgentTool”, Pesented at 12th Midwest Artificial Intelligence and Cognitive Science Conference, 2001.
    [3] Etzioni, O. and Weld, D. S., “Intelligent Agents on the Internet: Fact, Fiction, and Forecast”, IEEE Expert 10(4): 44–49, 1995.
    [4] Ferber, J., Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence, Addison-Wesley, 1999.
    [5] Fogarty, D. W., Blackstone, J. H., and Hoffmann, T. R., “Production and Inventory Management”, South-Western, 1991.
    [6] Frank, S. and Grasser, A., “Is It an Agent, or Just a Program?:A Taxonomy for Atonomous Aents”, Proceeding of the Third International Workshop on Agent Theories, Architecture, and Languages, Springer-Verlag(1996).
    [7] Hasse, R., “Building a Generic user Agent for Multi-Agent Integrated Enterprise”, Enterprise Integration Laboratory, University of Toronto (1997).
    [8] Huan, J. S. and James, T. L., and Trefor, P. W., ”Using Neural Networks to Predict Component Inspection Requirement for Aging Aircraft,” Computers Ind. Engng, Vol. 30,No. 2,257-267, 1996.
    [9] Huang, C.Y. and Nof, S.Y., “Autonomy and Viability-Message for Agent-Based Manufacturing Systems”, International Journal of Production Research, Vol.38, No.17, pp.4129-4148, 2000.
    [10] Miltenburg, J., Sparling, D., “Management and Reducing Total Cycle Time: Model and Analysis”, International Journal of Production Economics, Vol. 46, pp. 89-108, 1996
    [11] Shoham, Y., “ An Overview of Agent-Oriented Programming”, In Software Agents, ed J. M. Bradshaw. Menlo Park, Calif.: AAAI Press, 1997.
    [12] Steels, L. ”When Are Robots Intelligent Autonomous Agents?”, Journal of Robotics and Autonomous Systems, Vol. 15, No. 1-2, pp.3-9,1995.
    [13] Wooldridge, M., and Jennings, N. R., “Intelligent Agents: Theory and Practice”, Knowledge Engineering Review, 10(2), pp.115-152, January 1995.
    [14] Wooldridge, M., Sycara, K., & Jennings, N.R., “A Roadmap of Agent Research and Development”, Autonomous Agents and Multi-agent System, 1, pp.7-38, July 1998.
    [15] 葉怡成，「類神經網路模式應用與實例」，儒林圖書有限公司，八版，台北，2003年。

    無法下載圖示 全文公開日期 本全文未授權公開 (校內網路)
    全文公開日期 本全文未授權公開 (校外網路)
    全文公開日期 本全文未授權公開 (國家圖書館：臺灣博碩士論文系統)
    QR CODE