近十幾年來，全球航空運輸量不斷地成長，若飛航安全管理系統沒有重大的改進以應付日後龐大的交通流量，意外發生也很有可能會隨之增加。飛安事件的發生在統計機率的分析上是屬於稀有的事件，但是飛機失事對社會確會帶來極大的衝擊。 在航空運輸作業中如何管理風險與安全中是一個非常實際的問題，因此，在飛安風險管理上，必須作持續的改進，才能防避空難意外的發生。統計顯示接近70％以上的民航失事之導因為駕駛疏失(McFadden, 1993)，傳統的研究多著重在風險的概念性敘述和透過已發生的失事進行調查找出與人為疏失有關聯的因素。但較少研究是針對未來各種人為因素風險加以分析其長中短期趨勢，故航空公司和飛安監督單位難以研擬相關之防治措施。因此，本研究將建構一人為因素風險分析模式，期能預早發覺危機的發生，提供飛安監督單位或航空公司，作為決策與因應的參考，以提升飛安品質。
本研究利用比例危險模型（Proportional Hazards Model）與延伸型危險迴歸模型（Extended Hazard Regression Model）建構風險趨勢分析動態模型，從而建立各家航空公司針對各類事件的風險函數，並在考慮航空公司平日作業之飛安相關績效指標來評估與比較不同人為因素飛安事件在未來的風險值，若發覺某類人為因素事件可能發生風險機率過高狀況時，則可進行必要的措施或調整相關查核計劃來降低此類事件的發生風險，以達到風險管理的目的。

Risk and safety have always been of the first important issue in civil aviation, especially when the air transportation traffic volume grows rapidly and continuously. Although being a rare event in an absolute sense, aircraft accidents can have severe implications. A practical problem in air transport is how to manage risk and safety. In contrast to traditional reactive approach of casual analysis for finding linkage between human factors and history accidents, innovative proactive approach appears to be more essential for aviation risk management. Moreover, since the airlines’ operation performances actually reflect their safety ‘profile’ and overall safety quality, the relationship between human factors and safety performances should be considered while assessing the aviation risk.
The objective of this research is to develop an analytic method that uses both accident and safety data to quantify the aviation risk which are caused by human error. Assessment of the risk of aircraft accidents may be carried out in different ways, from highly intuitive to very formal and analytical.  The probability of an air accident is very low making it a difficult and complex task to proper explain, locate, and manage overall aviation safety. In this study we present a new methodology for assessment risk and safety in civil aviation. A specified proportional hazard model is investigated and demonstrated its applicability in aviation risk assessment.  Our approach takes into account the more complex relationships among relevant aviation risk factors.  The dependence of the aviation risk on airlines operational performance is also measured. Using the models presented, risk has been assessed as the probability of the occurrence of a specific type of human errors related aviation accident. So the potential human errors related risk is identified and monitored timely. The results can provide well references to the civil aviation communities to manage the aviation-safety risk, thus corrective action can be taken to reduce the happenings of aviation accidents.

LIST OF TABLES	III

LIST OF FIGURES	IV

NOMENCLATURE	V

CHAPTER 1 INTRODUCTION	1
1.1 Background and Motivation	1
1.2 The Goal and Methodology of the Research	2
1.3 Framework overview	3

CHAPTER 2 Related Work and Background Theory	4
2.1 Risk Assessment in Aviation	4
2.1.1 Researches related to accidents causing factors	4
2.1.2 Researches related to safety and risk assessment	9
2.2 Survival Analysis	14
2.2.1 Introduction	14
2.2.2 Terminology and Notation	14
2.2.3 Models Classification	16
2.3 Proportional Hazards Model	17
2.4 Extended Hazard Regression Model	20

CHAPTER 3 On the Use of Regression Model－ Proposed Solution	22
3.1 Preface	22
3.2 Data Analysis	23
3.2.1 Data Collecting Schema	23
3.2.2 Accident/Incident Record	24
3.2.3 Integrated Performance Indicators	25
3.3 Model Construction	31
3.3.1 Variables Description	31
3.3.2 Demonstration of Proportional Hazards Model	32
3.3.3 Demonstration of Extended Hazard Regression Model	33
3.3.4 Estimation for Parameters	35

CHAPTER 4 Results and Verification	39
4.1	Implementation Results	39
4.2	Models Verification	45

CHAPTER 5 Conclusions and Future Work	49
5.1 Conclusions	49
5.2 Suggestions for Future Work	50

Reference	51
LIST OF TABLES
Table 4-1 Potential risk exist in six airlines	39
Table 4-2 Probability of an accident within 10, 30 and 90 days	43
Table 4-3 Coefficients with ATOS covariates	44
Table 4-4 Coefficients with AW/OP/AF covariates	45
Table 4-5 Sample verification data set	46
Table 4-6 Summary of verification test for the sample data	47
LIST OF FIGURES
Figure 2-1 Hull Loss Accidents – Primary causes	7
Figure 2-2 The SHELL model	7
Figure 2-3 The “Swiss cheese” model of human error causation	8
Figure 2-4 Components of risk analysis	11
Figure 2-5 Framework for designing aviation safety studies	13
Figure 3-1 Process of model construction	23
Figure 3-2 Scheme of particular human factor accidents	24
Figure 3-3 AW/OP/AF based effecting factors	26
Figure 3-4 ATOS based effecting factors	26
Figure 3-5 Differences between the traditional system and ATOS	29
Figure 3-6 The Seven Airline Systems Defined in ATOS	30
Figure 4-1 Risk trend analysis estimated by EHR model	41
Figure 4-2 Risk trend analysis estimated by PH model	42
Figure 4-3 Risk trend analysis estimated by Poisson Process	42
Figure 4-4 Summary of verification test for the sample data	47

Abeyratne, R.I.R., “The regulatory management of safety in air transport,” Journal of Air Transport Management, 4 (1), 25-37, 1998.
Boeing Commercial Airplane Group, “Statistical Summary of Commercial Jet Aircraft Accident－Worldwide Operations, 1959-2004,” Seattle, 2005.
Bird, F., Management guide to loss control, Atlanta, GA, USA: Institute Press, 1974.
Braithwaite, G.R., Caves, R.E. and Faulkner, J.P.E.,“Australia Aviation safety－Observations from the ‘lucky’ country,”Journal of Air transport Management 4(1), pp. 55-62, 1998.
CAA, Civil Aviation Act of the Republic of China, Taiwan Civil Aeronautics Administration, 2005.
Ciampi, A. and Etezadi-Amoli, J., “A general model for testing the proportional hazards and the accelerated failure time hypotheses in the analysis of censored survival data with covariates,” Communications in Statistics- Theory and Method 79, pp. 651, 1984.
Cox, D.R.,“Regression models and life tables,” Journal of Royal Statistical Society 34, pp.187, 1972. 
Edwards, E. “Man and Machine: Systems for Safety,” Proceedings of British Airline Pilots Association Technical Symposium, London: British Airline Pilots Association, pp. 21-36, 1972.
Elsayed, E.A., Reliability Engeeneering, Addison-Wesley, 1996.
FAA,“Air Transportation Oversight System,” United States Department of Transportation, Office of the Secretary of Transportation, Washington, Report No. AV-2002-088, 2002.
FAA, “FAA Order 8400.10, Air Transportation Operations Inspector’s Handbook, appendix 6,”Retrieved from the world wide web, http://www.faa.gov/library/manuals/examiners_inspectors/8400/appendices/, 2006.
FAA, “Safety Reports-Aviation Safety Data Accessibility Study,” Federal Aviation Administration, Office of System Safety, Washington, DC 20591, 1997
ICAO, “Accident prevention manual,” ICAO document 9422, ICAO: Montreal, Canada, 1984.
Janic, M., “An assessment of risk and safety in civil aviation,” Journal of Air Transport Management (6), pp. 43-50, 2000.
Kaplan, S. and Garrick, B.J., “On the Quantitative Definition of Risk,” Risk Analysis, pp. 11-27, 1981.
Kleinbaum, D.G., Survival Analysis, Springer, 1996.
Lewis, C. et al., “System Safety: Accident Prevention Through Risk Reduction and Hazard Analysis,” Quarterly Journal of Flight Safety Information, fourth quarter, pp. 4-10, 2002.
Lowrance, W.W., Of Acceptable Risk, William Kaufmann, 1976.
McFadden, K. L., “An Empirical investigation of the relationship between alcoholand drup-related motor vehicle convictions and pilot flying performance,”Doctoral Dissertation, University of Texas at Arlington, 1993.
McFadden, K. L., “Comparing Pilot-Error Accident rates of male and female airline pilots,” Omega, The International Journal of Management Science 24(4), pp. 443-450, 1996.
McFadden, K. L., “Predicting Pilot-error Incidents of US Airline Pilots using Logistic Regression,” Applied Ergonomics 28, No. 3, pp. 209-212, 1997.
McFadden, K. L. and Towell, E. R., “Aviation Human Factors: A Framework for the New Millennium,” Journal of Air Transport Management 5, pp. 177-184, 1999.
Mineata, N., Avoiding Aviation Gridlock: A Consensus for Change, National Civil Aviation Review Commission, 1997.
PALISADE, Evolver, Retrieved from the world wide web, http://www.palisade.com.au/evolver/, 2006.
Reason, J., Human Error, New York: Cambridge University Press, 1990.
Rowe, W.D., An Anatomy of Risk, New York, NY: Wiley, 1977.
Sage, A.P. and White, E.B., “Methodologies for risk and hazard assessment: a survey and status report,” IEEE Transportation on System, Man, and Cybernetics SMC-10, pp. 425-441, 1980.
Satty, T. L., The Analytic Hierarchy Process, New York: McGraw-Hill, 1980.
Shyur, H.J., “A Semi-Structured Process for ERP Systems Evaluation: Applying Analytic Network Process,” Journal of e-Business 5(1), pp. 33-50, 2003.
Shyur, H.J., Elsayed, E.A. and Luxhoj, J.T., “A General Hazard Regression Model for Accelerated Life Testing,” Annals of Operations Research 91, pp.263-280, 1999.
Slovic, P., “Trust, emotion, sex, politics, and science: Surveying the risk-assessment battlefield,” Risk Analysis, 19(4), pp. 689-701, 1999.
Slovic P. and Weber E.U., “Perception of Risk Posed by Extreme Events,” Conference of Risk Management strategies in an Uncertain World, Palisades, New York, 2002.
United States Department of the Air Force, “Software technology support center’s guidelines for successful acquisition and management of software-intensive systems: weapon systems, command and control systems, version 2.0,”Management Information Systems, 1, USAF, Washington, DC, Ch. 1, 6 and 15, 1996.
Wiegmann, D. A. and Shappell, S. A., “Human error analysis of commercial aviation accidents: Application of the human factors analysis and classification system (HFACS),” Aviation Space and Environmental Medicine, 72(11), pp. 1006-1016, 2001.
Wong, J.T. et al., “Development of Risk Analysis and Assessment Model for Airline Safety,” Taiwan Civil Aeronautics Administration, 2004.