透過您的圖書館登入
IP:3.143.4.181
  • 學位論文

應用案例式推理於裝備維修系統之研究-以飛行模擬器為例

Case-Based Reasoning use in Equipment Maintenance Management System - A Case Study for Flight Simulator

指導教授 : 丁慶榮
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


飛行模擬器是現代飛行員訓練的必要裝備,由於模擬器量少且價格高昂,維持不易,實有必要建立維修專家系統,藉以蒐集以往的維修經驗,達到知識管理與組織學習的目的,加速故障診斷,降低維護成本。 案例式推理(Case-Based Reasoning, CBR)是近年來被廣泛運用的人工智慧方法之一,它是運用過去對一個案例解決過程的完整紀錄,來做為後續相類似問題產生時,提供解決問題之參考。 傳統的專家系統是採用的法則式(Rules-Based)推論法,隨著系統複雜度的增加,其所需的推論法則將成倍數成長,系統開發困難,且影響系統推論時間,若要完全涵蓋所有預估可能發生的狀況,其建置成本將是非常高昂,所以普及率並不高。 本研究應用案例推論(CBR)方法,提出情境模糊權重及機動門檻值,結合資料庫分層分類法、關鍵字法則,來建立飛行模擬器裝備維修管理系統,改善傳統案例式推理系統的效率問題,達到維修知識經驗的收集、累積、分享與學習。經由個案驗證,本研究所提出的案例式推理維修系統比傳統CBR方式,可以較快找到相似案例,未來可作為單位組織學習與知識管理的工具,協助模擬器裝備維修故障診斷及修護經驗之傳承。

並列摘要


Flight simulator is a very important training device for pilot training. Since simulator is very expensive, it’s maintenance cost is relatively high. Therefore, a good maintenance system is needed for technicians to record the reasons and actions for each maintenance case. Case-Based Reasoning (CBR) is a widely used method of Artificial Intelligence. CBR is able to utilize the specific knowledge of previously experienced, concrete cases for problem solving. A new problem is solved by finding a similar past case and reusing it in the new problem situation. This method is already used in many fields to develop expert systems. Traditional expert systems use IF…THEN rules to reasoning. When the system becomes more complicated, it will need more rules and times for reasoning. It costs very much and is hard to create a complex Rules-Based expert system. It’s why expert system isn’t so popular. This research applies the Case-Based Reasoning method that develops new attribute weights and moving threshold, with hierarchical classify database and keyword rules to create an equipment maintenance expert system for flight simulator. The results show that the CBR system can be used to collect maintenance experiences to share and learning by the technicians. The new system can quickly find similar case and provide good suggestions for maintenance jobs.

被引用紀錄


夏至緯(2009)。運用案例式推理於委外加工費用預測系統建構之研究 – 以DRAM產業為例〔碩士論文,元智大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0009-2807200914285800

延伸閱讀