失效模式與效應分析(failure mode and effect analysis, FMEA)廣泛運用於各種行業之可靠度及風險評估,以辨識產品或製程中風險較高的潛在失效模式 (potential failure mode, PFM)。傳統FMEA的風險評估,以順序尺度表示S、O、D之評估結果,因RPN計算結果範圍與S、O、D的定義域範圍不同,導致RPN判讀不易及計算結果恐有待商榷;模糊集合被提出以改善傳統FMEA評估的缺點,但是在RPN的運算過程中卻存在資料遺失的問題。本研究提出利用二元模糊語意表達模式 (2-tuple fuzzy linguistic representation model) 結合順序加權幾何平均 (ordered weighted geometric averaging, OWGA)運算子已獲得較合理之風險優先數 (risk priority number, RPN)的評估方法。考慮群體決策實務活動的過程,利用德菲法(Delphi method)以取得評估者對S、O、D的共同意見。最後,以一雲端運算服務之風險評估為案例,說明本研究所提出方法之實用性。
The failure modes and effects analysis (FMEA) approach is used to solving the reliability and risk assessment problems for identifying potential failure mode of the system. Existing proposals for solving the risk assessment assume that all of the parameters in the same definition doamin including severity(S), occurrence (O) and detectability (D), consequently it raised the doubtful computational results. Fuzzy sets were introduced to improve this difficulty, the results cannot be guaranteed due to the reason that it may lose of information in the computational process. In this study, the 2-tuple fuzzy linguistic representation model is proposed to determine the assessments of S, O, D and obtain the more reasonable results of risk priority number (RPN) with OWGA (ordered weighted geometric averaging) operator. Delphi method is adopted for aggregating experts’ opinions in the group decision-making process. Finally, an example of cloud computing service is demonstrated to illustrate the applicability of the proposed approach to risk assessment.