由於企業所面臨的環境快速的變動，所以企業要求軟體必須要有更高的回應速度、更高的品質、更具彈性的塑模流程以及更高的可維護性。而元件式軟體工程即是一種藉由整合現存的元件以生產更快速，更低成本及更高品質軟體的方法，且強調系統開發必須透過快速組裝那些屬性已經預先定義好且經過測試與驗證的元件來完成，所以元件軟體工程逐漸取代傳統的軟體工程，成為發展企業軟體的新一代軟體工程典範。
　　元件式軟體工程背後必須要有元件塑模方法論在背後支持，才能夠將企業的需求轉換成最終的元件，因此，元件塑模方法論的優劣，則對於元件式軟體工程有著顯著的影響。在經過調查後，發覺目前現有的元件塑模方法論大多存在著許多缺陷，如塑模方法停留在抽象的層次、缺乏明確的塑模準則、不具流程文件的連貫性以及無法滿足良好的元件塑模特性，所幸在調查過的方法論中，發現UML Components是一個符合良好方法論的特性且具有發展潛力的元件塑模方法論。
　　但是，UML Components塑模方法論本身仍然存在某些問題，如缺乏更明確的元件識別準則、更清楚的工具支援以及更精簡的塑模流程等。因此，本研究即針對UML Components方法論不足的地方，(1).利用[吳仁和2002]所提出的使用個案識別方法、[Marquis2002]所提出的”Fan-In”與”Fan-Out”的觀念以及使用個案關係轉換法則擴充改善了元件塑模準則的問題；(2).利用資料辭彙、資料關係矩陣以及操作參照矩陣來改善元件塑模工具的問題；(3).針對原本元件塑模流程不合理之處，重新設計其流程模型，產生一個以需求定義為出發點，依序進行元件識別及元件互動分析，最終產生元件規格定義的四階段元件塑模流程模型。
而本研究則透過改善上述三個缺點，提供元件式系統分析人員進行元件系統塑模時有依偱的標準，另外幫助系統開發人員能夠更清楚準確的發展出滿足企業需求的系統以及保障元件的品質，並加快軟體開發的速度。

Due to the fact that the environment which business confront is very dynamic, so the business require the software to reduce the response time, provide the high quality, and supply the flexible modeling process and high maintainability. In order to meet the requirement, a new software solution is emerged. Component-based software engineering is a method to integrate existed component to produce faster speed, lower cost and higher quality software. And it is concerned with the rapid assembly of systems from components where components and frameworks have certified properties; and these certified properties provide the basis for predicting the properties of systems built from components. Therefore, component based software engineering replace the position belong to traditional software engineering progressively, it become the next generation software engineering paradigm.
Because businesses want to transform their requirement to be final components, so the component-based software engineering must have component modeling methodology in the system analysis stage. For this reason, the advantage and disadvantage of the component modeling methodology makes a very deep influence on component-based software engineering. And after the survey we made, we discover the greater parts of the existed component modeling methodology have a lot of defects, for example: the modeling method stay at abstract level, lack of clear and definite modeling rules, short of consistence of process document and can’t meet the good component modeling characteristics. Fortunately, in our surveyed methodology, we discover the “UML Component” is a component modeling methodology which can meet the good component modeling characteristics and have the develop potential.
But the “UML Component” methodology still has some problem, for example: lack of explicit component identify rules, complete tools support and simply modeling process. Due to these reasons, our study provides three solutions to refine and extend the “UML Components” methodology. First of all, we utilize the use case identify method by [吳仁和2002]、the concept of “Fan-in and Fan-out” by [Marquis2002] and the use case association transform method to refine the problem of component modeling rules. Secondly, we employ data glossary, data relation matrix and operation reference matrix to improve the problem of component modeling tools. Finally, we redesign the process model which includes four stages as follows, requirement acquisition, component identification, component interaction and component specification to resolve the irrational parts of original modeling process.
With this refined methodology, the system analysts can follow the stand modeling rules and process. In the other hand, the system developer also can develop the system which meet the business requirement exactly, ensure the component quality and speed up the software develop rate.

第一章 緒論 1
1.1　研究背景與動機 1
1.2　研究目的與範圍 2
1.3　研究流程 2
第二章 文獻回顧 3
2.1　元件塑模方法論 3
2.1.1　Catalysis 4
2.1.2　SCIPIO 7
2.1.3　O2BC 10
2.1.4　UML Components 12
2.1.5　綜合比較 18
第三章 元件塑模方法論 20
3.1 概述 20
3.2 需求定義階段 22
3.2.1　找出行為者 22
3.2.2　找出使用個案 22
3.2.3　描述使用個案與關係 23
3.2.4　產生使用個案模型 24
3.3 元件識別階段 25
3.3.1　識別企業型態模型及企業介面 25
3.3.2　識別系統介面及操作 29
3.3.3　產生初步元件規格及架構 30
3.4　元件互動階段 33
3.4.1　發掘企業操作 33
3.4.2　修飾介面及操作 37
3.4.2.1　維護參照完整性 37
3.4.2.2　整理企業介面與系統介面 38
3.5　元件規格階段 39
3.5.1 專有名詞 39
3.5.1.1 前置條件與後置條件 39
3.5.1.2 物件控制語言 40
3.5.1.3 介面資訊模型 41
3.5.2　定義介面資訊模型 42
3.5.2.1　企業介面資訊模型 42
3.5.2.2 系統介面資訊模型 43
3.5.3 定義操作前置與後置條件 45
3.5.4 定義元件介面限制 47
3.5.4.1　定義提供與使用之介面限制 47
3.5.4.2　定義介面互動限制 48
第四章 元件塑模方法論實例 50
4.1　需求定義階段 50
4.1.1　找出行為者 51
4.1.2　找出使用個案 51
4.1.3　描述使用個案與關係 52
4.1.4　產生使用個案模型 57
4.2　元件識別階段 58
4.2.1識別企業型態模型及企業介面 58
4.2.2　識別系統介面及操作 67
4.2.3產生初步的元件規格架構 73
4.3　元件互動階段 76
4.3.1　發掘企業操作 76
4.3.2　修飾介面及操作 81
4.3.2.1　維護參照完整性 81
4.3.2.2　整理企業介面 83
4.4　元件規格階段 85
4.4.1　定義介面資訊模型 85
4.4.1.1　企業介面資訊模型 85
4.4.1.2 系統介面資訊模型 89
4.4.2 定義操作前置與後置條件 93
4.4.3 定義元件介面限制 114
4.4.3.1　定義提供與使用之介面限制 114
4.4.3.2　定義介面互動限制 115
第五章　結論與建議 118
5.1　回顧與探討 118
5.2　未來研究方向 119
參考文獻 121

[吳仁和2002]　吳仁和、林信惠，系統分析與設計，智勝，台北，2002。
[Bachman2000]　Bachman, F., Bass, L., Buhman, C., Comella-Dorda, S., Long, F., Robert, J., Seacord, R., Wallnau, K.,” Volume II: Technical Concepts of Component-Based Software Engineering”, at URL: http://www.sei.cmu.edu/pub/documents/00.reports/pdf/00tr008.pdf, 2000
[Barn98]　Barn, B, Brown, A., W., “Methods and Tools for Component Based Development”, Technology of Object-Oriented Languages and Systems, August 1998
[Bergner99]　Bergner, K., Rausch, A., Sihling, M., Vilbig, A., “Componentware - Methodology and Process”, The 21st International Conference on Software Engineering, 1999
[Booch94]　Booch, G., Object-Oriented Analysis and Design-With Applications, Addison-Wesley, USA, 1994
[Chessman2001]　Chessman, J., Daniels, J., UML Components, Addison-Wesley, Boston, 2001
[D’Souza99]　D'Souza, D., Wills, A. C., Catalysis: Objects, Frameworks, and Components in UML, Addison-Wesley, Boston, 1999
[Ganesan2001]　Ganesan, R., Sengupta, S., “O2BC: a Technique for the Design of Component-Based Applications”, Proc. 39th International Conference and Exhibition on Technology of Object-Oriented Languages and Systems, Santa Barbara, USA, 2001
[Hallsteinsen99]　Hallsteinsen, S., Skylstad, G., “The Magma approach to CBSE”, The 21st International Conference on Software Engineering, May 1999
[Heineman2001]　Heineman, G., T., Councill, W., T., Component-Based Software Engineering-Putting the Pieces Together, Addison-Wesley, USA, 2001
[Sang99]　Sang, D., L., Young, J., Y., Eun, S., C., Soo, D., K., Sung, Y., R., “COMO: A UML-Based Component Development Methodology”, Proc. 6th Asia Pacific Software Engineering Conference, Japan, December 1999
[Veryard98a]　Veryard, R., “SCIPIO: Aims, Principles and Structure”, SCIPIO Consortium, April 1998
[Veryard98b]　Veryard, R., “SCIPIO: Effective management of component-based system project”, SCIPIO Consortium, November 1998
[Marquis2002] Marquis, G., P., “Application of traditional system design techniques to web site design”, Information and Software Technology, Vol. 44, No. 9, pp.507–512, 2002
[Flowler99]　Fowler, M., Scott, K., UML Distilled Second Edition, Addison-Wesley , Boston, 1999
[Kwon99] Kwon, O. C., Yoon, S. J., Shin, G. S., “Component-Based Development Environment: An Integrated Model of Object-Oriented Techniques and Other Technologies”, The 21st International Conference on Software Engineering, Los Angeles, USA, 1999
[Warmer99] Warmer, J., Kleppe, A., The Object Constraint Language-Precise Modeling with UML, Addison-Wesley, 1999