多站平行佈置流程式彈性裝配系統之動態可靠度探討

摘要

對現代產業而言，彈性製造系統與彈性裝配系統的引進已成爲產業界爲適應市場競爭、提高產品品質、及降低日以增加之直接人工成本的重要生產設施。對於使用彈性裝配系統的高度自動化裝配廠而言，系統之可靠度直接影響其生產效率與產品之品質甚鉅。故如何正確預測彈性裝配系統之動態可靠度，乃成爲一重要之課題。由於系統之複雜性，使得彈性裝配系統之系統可靠度問題，至今仍未被有系統的予以探討，而相關之研究則多局限於彈性製造系統與單一機器人之可靠度分析。因此，本研究乃對於彈性裝配系統之系統可靠度問題，首先以馬可夫程序模式來描述其系統動態可靠度與妥善率，並以類神經網路來建立系統之動態可靠度預測模式，隨後透過例證之結果說明類神經網路爲一種準確度良好之可靠度與妥善率之預測工具。

Abstract

The reliability problems of FASs have not been discussed systematically due to the system's inherent complexity. Most of the related researches limited their focuses on the Flexible Assembly Systems (FASs) or robot reliability analysis. In this research, we first describe the dynamic reliability and availability model of FASs through markov process. Second, a reliability prediction model of discussed problem is constructed by applying neural network technique. Finally, a case study is demonstrated to verify that the designed neural network model is a useful tool for predicting the reliability and availability of FASs, and the statistical analysis is also provided.

關鍵詞:

• 彈性裝配系統
• 動態可靠度
• 妥善率
• 馬可夫程序
• 類神經網

Keywords:

• Flexible Assembly Systems
• reliability
• availability
• markov process
• neural network