本研究嘗試使用不同方法進行飛航數據例行操作評估(routine operational measurement)以辨識不穩定進場的風險，主要使用監控參數為：指示空速選擇(Indicated airspeed selection)和下降率(descent rate)。針對例行操作評估，本研究進行四種分析，包含：飛航參數數據圖表分析、離群值分析、穩定進場條件檢核、及變異數分析等。上述分析可辨識離群值的航班，以及可能產生問題的場站和跑道，然後針對這些問題航班檢視其開始下降全程飛航數據以釐清事件的成因。值得注意的是，完整的飛航數據風險辨識分析應包含：例行操作評估分析、飛航數據檢視與事件調查、與個案航空公司討論，必要時進行改正措施並觀察改正措施的有效性；本研究目前進展只涵蓋了前三個階段。分析結果顯示，有些經由例行操作評估辨識出的問題航班，可能沒有被超限分析(exceedance analysis)識別出來，抑或有被識別出來然而卻被忽略。因此例行操作評估可以被視為風險辨識額外的保護措施。
根據本研究的分析結果，建議將所有識別出的航班建檔存入資料庫，並標記每個識別航班的相關屬性；當資料庫中的航班累積至一定量時，風險分析和評估將會更精準且具說服力。本研究雖然致力辨識所有可能產生疑慮的航班，然而分析結果顯示大部分的航班都屬正常。即使有些航班於操作上有些許瑕疵，但現況並未產生致命失誤，不過從起司理論觀點，當與其他狀況同時發生，或許將導致不良結果，因此本研究建議應更嚴謹檢視這些有瑕疵的航班，從風險的觀點討論，儘量避免忽視其中所隱含的訊息，至少將這些航班儲存至資料庫，以供未來風險辨識參考。

This study explores ways to conduct routine operational measurement (ROM) of flight data to identify risk for unstable approach. Indicated airspeed selection and descent rate are two main parameters for monitoring. This study conducts four types of analyses: flight parameter profile analysis, outlier analysis, stable approach criteria, and ANOVA analysis. This approach is helpful for identifying outlier flights, question aerodromes, or runways. A thorough flight data examination would be conducted to investigate the reasons behind the event. It is noted that the entire loop includes ROM analysis, event investigation, discussion with Airlines, remedies developed if necessary, and observation of effectiveness of remedies. The progress of this study covers first three phases only. Some flaws, identified by ROM analysis, may not be detected or be detected but ignored by exceedance analysis. ROM analysis could be another protection to supplement such a deficiency for risk identification.
It is recommended to store the identified cases in a database for risk analysis and assessment. Even though this study tries very hard to identify some flights with questions, it may end up that most are normal flights. Some may have flaws, but at this stage, not fatal mistakes. It is recommended not to overlook any message delivered by these flaw flights easily. And at least, store them in a database for risk analysis.

1.International Civil Aviation Organization (2014), “Manual on Flight Data Analysis Programmes (FDAP)”.
2.International Civil Aviation Organization (2013), “Safety Management Manual”
3.Federal Aviation Administration (2004), “Flight Operation Quality Assurance”.
4.Civil Aviation Authority (2013), “Flight Data Monitoring”.
5.Civil Aeronautics Administration (2016), “Civil Aviation Act”.
6.Civil Aeronautics Administration (2017), “Aircraft Flight Operation Regulations”.
7.Civil Aeronautics Administration (2014), “Safety Management System”.
8.Civil Aeronautics Administration (2015), “Safety Performance Indicators”.
9.Civil Aeronautics Administration (2017), “Airliner FOQA Criteria”.
10.International Air Transport Association (2011), “Runway Excursion Analysis Report 2004-2009”.
11.European Aviation Safety Agency (2012), “Review of Accident Precursors for Runway Excursions”.
12.European Aviation Safety Agency (2014), “Study for Runway Excursion Precursors”.
13.Stolzer, A. J. (2003), “Fuel consumption modeling of a transport category aircraft: A flight operations quality assurance (FOQA) analysis,” Journal of Air Transportation, 8(2), 1-18.
14.Li, L., Hansman, R. J., Palacios, R., and Welsch, R. (2016), “ Anomaly detection via a Gaussian Mixture Model for flight operation and safety monitoring,” Transportation Research Part C: Emerging Technologies, 64, 45-57.
15.Wang, L., Wu, C., and Sun, R. (2014), “ An analysis of flight Quick Access Recorder (QAR) data and its applications in preventing landing incidents,” Reliability Engineering & System Safety, 127, 86-96.
16.Yeh, Y.P.(2004), The Key Factors of a successful FOQA System, National Cheng Kung University,
17.Yao, C.W.(2006), The Risk Analysis of Exceeding Tailwind-Limitation Approach, National Cheng Kung University,