關鍵詞：協力式設計系統、資料庫、黑板式專家系統。

論文提要內容： 鑒於傳統程序性程式在土木工程界中已廣泛地應用，又專家系統也成功的在結 構設計分析上顯現其功能及效益。故本研究便嘗試結合程序性程式與專家的經驗性 知識，形成共同合作的解題模式，成為協力式設計系統，期望能更有效率地進行結 構設計與分析的工作。

　　傳統的程序性程式在解決繁雜大量的數值問題時，有其快速與精確的優點；專 家系統則是擷取專家的經驗性知識解決問題。然而，這兩者基本架構的不同（例如 ：設計理論、資料結構、以及執行語法等的不同）常使它們發展成為兩種截然不同 的系統，而且這兩種獨立系統在資料的交流上也不是很理想。所以本研究以黑板式 專家系統架構整合知識庫系統與程序性程式的解題模式，其作法是採數值運算程式 （以FORTRAN與C語言寫成）植入以人工智慧語言所建立的黑板設計架構當中。

　　本研究所發展的協力式設計系統是建在HP工作站上，並以人工智慧軟體發展環 境KEE（Knowledge Engineering Environment）來建立系統架構。並以平衡式懸臂施工 橋的設計領域建立設計知識庫與FORTRAN及C程式來測試黑板系統之可行性。系統的 特點在於其明確的知識表示法、機會式的解題方式與獨立的知識表示架構（包括數 值運算與經驗性知識的分離以及控制與領域知識的分離），期望能成為一般性的共 同設計架構，以最有效的方式幫助工程師解決設計問題。

Keywords:WWW, Database, Knowledge-Base Expert Systems, Construction Management, Engineering Information Management.

Algorithmic programs are best for solving large scale numerical problems because of its high efficiency and accuracy. Knowledge-based expert systems capture the heuristic knowledge that experienced designers may have. They both are very important tools for structural design. However, the fundamental differences between procedural programs and expert systems (eq., design theories, data structures, implementation languages, etc.) often make them separate systems; and the communication between them is usually poor.

This thesis presents a collaborative structural design environment that integrates knowledge-based and algorithmic problem solving methods. The approach is to embed procedural program (written in FORTRAN and C language) in a blackboard design framework. In the blackboard model, algorithmic or heuristic knowledge needed to solve a problem is partitioned into independent knowledge sources that are grouped into several knowledge modules in the knowledge base. The knowledge sources modify only a global knowledge structure (the blackboard) and respond opportunistically to the changes on the blackboard. Heuristic and algorithmic knowledge is declaratively represented in the system using an object-oriented (frame-based) approach.

The collaborative design framework is implemented in Intellicorp's KEE (Knowledge Engineering Environment) on a HP workstation. A domain knowledge base as well as FORTRAN and C programs for balanced cantilever bridge design were built to test and evaluate the collaborative design system.