飛航安全風險評估模式之構建

RISK ASSESSMENT MODEL ON FLIGHT-SAFETY MANAGEMENT

飛安風險評估乃是一種運用有限資源以較小的成本或較有效的方法，來達到減少飛安事故及其損失的過程，包括風險的衡量與分析兩部分。若我們在日常即能針對各項影響飛安的風險因素嚴加監控，當這些因素呈現故障或缺失狀態超過設定的門檻時即能發出警訊，告知風險管理者採取適當的處理措施，將可防範飛安事故於未然。由於傳統 FMECA的最大缺點即在其 RPN是由各項衡量因子的衡量值相乘而得，其易造成衡量值相主抵銷的盲點，故其運用在安全性訴求極高的航空產業並不適宜。為此，本文乃利用座標組合法和歐基里德距離原理來發展一套組合衡量值的風險衡量模式，再參考 ALARP風險概念來進行風險分析模式之構建。透過此飛安風險評估模式的運作，風險管理單位便可建立彈性化的即時風險控管系統。首先經由專家問卷調查蒐集各項飛安風險 因素之內生衡量因子（包含因素重要度、故障危害度和故障難檢度等）之衡量值後，接著利用 k-標準差的統計技術來繪製風險分析圖，並針對國內影響飛安中的機械因素進行實證探討。風險管理單位透過本模式的操作可發揮預防診斷的功能，且隨著對風險分析圖中之下上界風險水準門檻（即調整 k參數）的設定，可達到有效監控各項風險因素之風險水準現況，並及早採取相對應的風險處理措施，儘速發現異常的故障或缺失狀況，以避免後續重大飛安事故的發生。本文以量化分析方式處理飛安風險衡量問題，除有別於採取原則性質化風險分析之研究外，更能讓安檢數據說話，當可提供航空業界相關單位之參考。

An efficient flight-safety risk assessment model using the lowest cost can reduce flight accidents and the loss. The model includes two parts, risk measurement and risk analysis. If the flight-safety risk factors can be strictly monitored and the appropriate measures can be adopted while failure occurred, flight accidents will be greatly decreased. Due to the serious defect of using the RPN, which was calculated from the product of the scores of measurement factors, the traditional FMECA principle is not applicable to the risk analysis of aviation industry requiring much higher safety concern. Thus, we developed a model applying the coordinate combination method and the Euclidean distance formula to combine the scores of measurement factors to measure the risks. Also, we modified the ALARP approach to establish the risk analysis model. Then the operating of the risk analysis diagram (RAD) was interpreted. That is, an elastic monitoring/controlling mechanism would be established to timely monitor the status of each risk factor. Hence, we first obtained the scores of each endogenous measurement factors of each risk factor screened from the expert’s questionnaire survey. Next, we employed the statistical k-σ technique to draw the RAD for each risk factor. A practical study for technological risk factors of flight-safety was presented in this paper. By operating this model, the risk manager can appropriately adjust the k parameter to set up the upper-bound threshold of level of risk to efficiently monitor the risk level and adopt the corresponding measures of risk treatment to find out the extraordinary hazards as early as possible. In this study, the quantitative risk analysis method was used to measure and cope with the risks. This is different from the conventional manner of the qualitative risk analysis. So, the contributions of this study may provide help for the related departments in civil aviation to manage the risks of flight-safety.