本研究整合了多個物件導向程序與元件程序的優點，提出一個以元件式的系統開發思維為中心、使用系統化的元件再用映對，強調整合性與再用性的元件式開發程序。本研究所提出的程序以Unified Process為基礎，利用UML為塑模語言，輔以相關元件式系統的開發經驗，同時結合其他物件或元件式系統的開發方法的優點，包括：Select Perspective Process、IBM San Francisco Roadmap Process、XP等。在此程序中，明確定義出不同的開發階段，一連串的彼此相關的開發活動，以及各開發活動應該如何完成的指引原則，為元件式軟體開發提供參考；最後並以郵務先導系統的開發產出，作為此程序的實例說明。

知識是組織中最重要的資源，需要管理方能發揮更大的效益，以提昇組織的競爭優勢。所謂知識管理，就是使既有的知識產生作用，亦即有系統、有組織地應用知識，進而創新知識。在網際網路時代，電子圖書館的發展日漸蓬勃，提供知識流通與利用的良好環境。知識社會中的圖書館除了是知識寶庫之外，也應該積極地使知識發揮更多的作用。本研究即希望應用知識管理的觀念，建立一個具有三項特色的開放式電子圖書館學習與管理的環境：1.提供讀者一個互動式的學習環境；2.讀者可以協助參與管理的工作；3.讀者亦可分享與回饋知識給電子圖書館。 首先，探討電子圖書館知識管理基本架構的主要組成元件，包括：知識工作者、知識倉儲與知識活動，並將知識活動予以流程化，提出知識管理程序。其次，電子圖書館知識管理系統的建置，乃是藉由物件導向軟體工程的系統開發方法，將針對知識工作者的角色分類與知識代理人的模組定義，並探討知識管理程序的使用個案模式與知識倉儲關鍵物件之定義。最後，建立電子圖書館知識管理系統的分析模式，作為電子圖書館知識管理系統雛型建置的基礎。 / constructing a knowledge management environment in digital libraries

處在消費者需求快速變動的時代裏，製造業必須即時取得銷售資訊以掌握消費趨勢，並透過生產的製程調整以因應消費者需求的變動；而快速回應(Quick Response, QR)則被視為能夠藉由資訊科技的使用，降低產業經營成本，並即時回應消費者需求，提升產業競爭優勢的利器。因此，本研究之目的乃是希望在供應鏈成員能夠攜手合作（如透過策略聯盟的方式）的前提下，藉由此製造業快速回應系統架構及模式的引進，達到資訊流垂直整合的目的，並有效地傳遞產銷資訊，使供應鏈成員能針對消費者需求擬定更好的生產及銷售策略，進而達成消費者、零售業、批發業與製造業四贏的局面。 本研究透過文獻探討的方式，提出一個滿足快速回應理念的製造業資訊系統架構，並對其功能加以說明；其次，再透過整合型物件導向系統分析方法，建構出該系統之物件模式，並描述物件間的關係；最後，再以程式語言撰寫資訊垂直整合單元的雛型系統，以驗證本研究之可行性。 本研究的貢獻在於提出一個製造業快速回應系統的架構與物件模式，使製造業可以在建構此類的系統時，有一明確可依循的參考架構，以加速該系統之建置、維護，與修改。 / In the age of rapid changes in customers' needs, the industry must get the sales information on time, and then use this information to adjust its production processes to meet the changes. Quick response systems use information technologies to reduce the operating cost, to respond the customers' needs and changes on time, and to gain the advantages in a business competition. Under the assumption that all members of the supply chain would have cooperated (for example, the members can be cooperated by strategic alliance), the purposes of this research is to develop an architecture and an object model for industrial quick response information systems. Through the usage of this architecture and this model, we hope to meet the purpose of a vertical information integration among the members in the chain; so the members can share their information, and draw up the better production and sales strategies based on the customers' needs. Hence customers, retailers, wholesalers, and manufacturers can reach the win-win situation. Through the literature survey, we first try to propose an architecture for manufacturing information systems with the concept of quick response. Secondly, we use the integrated object-oriented approach to construct an object model for the proposed architecture and to describe the relationships among the objects in the model. Finally, we develop a prototype to show that how vertical information integration can be attained in this architecture. The major contribution of this research is to raise an architecture and an object model for industrial quick response information systems. Manufacturers or manufacturing software industries can use this architecture and this model as a reference to develop their owned systems. We believe that this kind of reference could speed up the time of the system development, maintenance and modification.

在電腦整合製造（Computer Integrated Manufacturing）的系統中 ，資訊整合的良好與否，往往是系統成敗的關鍵所在。在現行組織中，常 常因為資訊的缺乏相互溝通，造成資訊的重覆製作，遂使得人力、時間經 常耗費在一些沒有價值的活動上，因而造成組織中的工作效率無法提升； 然而要真正瞭解組織中各個地區、各部單位到底擁有那些資訊，以及需求 那些資訊，並且將這些往來的資訊整合起來，卻是件相當不容易的事。為 達到此目的則端賴一完整的資訊流程（Information Flow）分析方法，並 參照電腦整合製造資訊流程參考模式（Reference Model ）來幫助我們尋 找及建立製造企業內正確的資訊流程。在本研究中，我們採用由上往下 （ Top-Down）的分析方式，並透過功能拆解（Functional Decomposition）的技術來分析電腦整合製造系統中資訊流程的關聯性， 最後再使用物件導向方法（Object-Oriented Method）來分析系統內細部 的資訊流程；經上述方式，我們發展出一套電腦整合製造系統資訊流程分 析方法，此方法中包含了分析進行的步驟與使用工具。此外，我們亦將依 據此套分析方法，試圖提出一套電腦整合製造資訊流程參考模式，以提供 分析時參考的依據，並驗證此套分析方法的可用性。

近年來由於全球資訊網的盛行，造成網際網路上的資料需求量大增，資料庫管理系統與全球資訊網必須結合以滿足客戶端的資訊需求。但隨著網路異質性的增加，不同的平台、作業系統及通訊協定不斷地加入使用，不一致的問題亦隨之而來。 　　同時，物件資料庫的兩大標準，SQL3與ODMG，在許多觀念上不儘相同。因此，本研究嘗試從物件模型及查詢語言兩方面著手，以物件觀點來分析兩者的相同及相異處，提出一對應模式，並實做一個資料庫中介系統，以轉換兩種語言在語法上的差異。 / In recent years, the complexity of database systems has been enhanced under the development of client/server architecture and distributed computing systems. Usually this type of system combines different hardware, network protocols or DBMSs. 　　Because the Internet and WWW are more and more popular, many people regard "Network as a computer" or "Network as a global database." It is obvious heterogeneous databases will be connected via WWW in order to provide more information. In the meantime, new OODBMS standards, SQL3 and ODMG, are emerging. Although the two standards both support for object facilities, they are quite different in object model and query languages. Therefore, the mapping between these two standards is necessary. We propose a comparison model and develop an experimental gateway according to the model.

傳統會計與決策管理系統受限於現行財務會計報導的架構，並未儲存與財務無關，或不能以金額衡量的資訊，無法對非財務性決策提供有效的幫助。而資訊語意的差異更使得傳統會計與決策管理系統不易與企業內其他資訊系統互通，造成系統整合上的困擾，歷來不斷有學者探討會計與決策管理系統資訊模式建立之方法，以增進會計與決策管理系統資訊模式的有用性。 本研究提出會計與決策管理系統資訊模式建立之框架與方法，強調會計與決策管理系統之建構，應與企業策略發展整體考量，同時導入REA model整合企業整體資訊模式與財會專業資訊模式，並採用UML作為系統發展流程之塑模工具，編製系統發展有關文件，以利後續選擇解決方案的映對程序，確認解決方案符合使用者需求，及增進會計與決策管理系統在企業內的整合性與互通性。 / Traditional accounting and decision management information system cannot support non-financial decision making, because there is only stored financial information without non-financial information due to the current financial accounting reporting structure. The information semantic differences block interoperability between accounting and decision management information system within the organization. And it also has caused the challenge of enterprise information integration. Researchers strive to find viable information modeling process or method to help accounting and decision management information system to be constructed. This thesis presents accounting and decision management information system information modeling framework and process. The framework stresses the construction of accounting and decision management information system must connect with enterprise vision and strategy. It also adopts and expands REA model to integrate accounting information model and enterprise information model at information modeling process. UML is a high-level modeling tool to develop system processes. The goal is to provide a standard method to develop system blueprints beyond tool-specific system solutions and to apply UML modeling documents to map different system solutions and make sure the solution fit with user requirements, and improve the integration and interoperability for accounting and decision management information system design and development.

隨著網際網路的蓬勃發展，傳統的地理資訊系統也從早期的單機系統朝Web GIS的方向發展，但是關於地理資料的格式卻仍舊莫衷一是，沒有一個為業界所共同遵從的標準，其結果除了人力與財力資源的浪費，也造成不同單位間的地理資料整合上的困難。傳統的電子地圖採用點矩陣的圖形格式，因檔案太大而常造成資料傳輸上的延遲，而且圖形的解析度與品質不佳；此外在建置上十分不便，需要專門的處理軟體才能進行編輯修改。因此傳統GIS的處理方式已經不能滿足網路化的GIS應用系統的資訊需求。 　在XML出現之後，W3C與業界也體認到這樣的需求，因此新一代的網路二維圖形標準，可變動向量圖形（Scalable Vector Graphics，SVG）便應運而生，雖然目前SVG仍在候選建議階段，但鑑於它的諸多特點與優越性，目前已獲得業界各大公司的支持。本論文便在此種情況下產生研究動機，我們期望透過SVG這種新興的向量圖形格式，找出解決目前GIS在地理空間資料上所面臨的困境。經過本研究所實作出來的雛形系統顯示，以SVG作為新一代的Web GIS地理空間資料格式不但可行，而且突破許多傳統方式的瓶頸和缺點。 / With the development of Internet, traditional GIS was being changed from single user to Web GIS. But the file formats of geographic spatial data are still various. We don’t have a simple and effective standard. The result are financial and manpower resource wasting. It is inconvenient and difficult to integrate geographic spatial data between different GIS and organizations. Traditional maps are bitmapped format, with lower quality at high resulation. We have to use particular tools when modifying bitmaps. Therefore, traditional way in GIS should be improved. 　The Scalable Vector Graphics（SVG） format is a new XML grammer for defining vector-based 2D graphics of the Web and other applications. SVG was created by World Wide Web Consortium（W3C）, the non-profit, industry-wide, open–standards consorium. Over twenty organizations, including Sun Microsystems, Adobe, Apple, IBM, and Kodak, have been involved in defining SVG. SVG is a subset of XML, which is rapidly becoming the foundation for all modern Web applications. SVG gave us a motive to do the investigation. We hope to use the SVG format to solve the problems in GIS. In the result of the prototype system in our thesis, we find out that SVG should be the better, effective spatial data format of Web GIS.

知識時代的來臨，伴隨著開放式資訊環境的發展，個人、團隊間之知識共享環境以及衍生出之知識貿產，已衍然成為企業組織生存與成敗的關鍵因素。因此，本研究根據Porter之價值鏈模式，將其“線性”模式作適當地調整，用以配合“網狀式”的知識分享模式，並且融人Nonaka之組織知識創新過程，建立了企業「知識價值鍊」模式，以表達知識分享與創造的動態關係。而為落實知識管理機制於企業日常作業中，本研究提出了新的資訊系統分析過程，係以文化的觀點分析組織知識結構─文化知識，納入知識管理系統分析與設計的考量中。因此，本研究針對企業內不同階層的知識工作者即文化團隊，提出標準模組化的物件：「知識代理人」，其中，代理人的三層架構代表了不同次文化團隊的需求。 知識管理系統的建置，乃藉由物件導向軟體工程的系統開發方法（如使用個案模式），結合新的資訊分析過程，作為本系統模式發展的基礎。故知識管理系統主要涵蓋兩種分析構面，用以建構整體的知識管理系統，一為代表組織知識管理結構的知識代理人，另一則是表示企業日常營運活動的物件模式。而在研究過程中以行銷公司為例，建置系統分析雛型來加以探討與印證。 / The advent of a knowledge age, the critical factors of a survival and successful business are knowledge sharing and derivative knowledge assets between individual and group in an open informational environment. Consequently, our research based on Porter's '" value chain model J and adapted his "linear" model for "network" knowledge sharing model. And then the adapted model integrated with Nonaka's 「organizational knowledge creating process」 to construct the 「knowledge value chain」 model of a business. So the model can express the dynamic relations of sharing and creating knowledge. For implanting the knowledge management mechanism in business operational activities, we proposed new analysis processes of an information system as a knowledge management system. The method analyzes organizational knowledge structure in a culture aspect as culture knowledge, and subsumes the aspect under information system analysis and design. According to the different level of the knowledge workers as cultural groupings, we proposed standard model objects as knowledge agents. Interalios, the three levels of agents represent different demands in different context groups. During constructing knowledge management system, we used object-oriented software engineering in system developing method, for example the use case model. Therefore, the knowledge management system contain two analysis dimensions. One is the knowledge agent which represents organizational knowledge management structure. The other is the object model which displays business daily operations. In our research process, we illustrated the knowledge management system prototype with the marketing corporation.

The advent of a knowledge age,the critical factors of a survival andsuccessful business are knowledge sharing and derivative knowledgeassets between individual and group in an open information environment.Consequently, our research based on Porter's "value chain model" and adaptedhis "linear" model for "network" knowledge sharing model. And then theadapted model integrated with Nonaka's "organizational knowledge creatingprocess" to construct the "knowledge value chain model" of a business.So the model can express the dynamic relations of sharing and creatingknowledge.For implanting the knowledge management mechanism in business operationalactivities, we proposed new analysis processes of an information system. Themethod analyzes organizational knowledge structure in a culture aspect asculture knowledge, and subsumes the aspect under information system analysisand design. According to the different level of the knowledge workers as culture groupings, we proposed standard model objects as knowledge agents. Interalios, the three level of agents represent different demands in different context group. During constructing knowledge management system,we used object-oriented software engineering in system developing method,for example the use case model.Therefore, the knowledge management system contain two analysis dimensions.One is the knowledge agent which reprents organizational knowledge managementstructure. The other is the object model which display business daily operations.In our research, we illustrated the knowledge management system with the marketing corporation.

專家決策支援系統（Expert Decision Support System）是一種結合專家 系統與決策支援系統功能於一體的電腦化資訊系統；它同時具備有處理定 性資料與定量資料的能力，可以進行各種知識法則的推論及數量模式的執 行與分析，以完整的支援問題之求解過程。 / The expert decision support system(EDSS) is a computer-based nformation system which integrates functions of the expert systemnd the decision support system. EDSS, which is capable ofrocessing symbolic and numerical data, provides the