Font Size: a A A

Research On Techniques Of Risk Management For Airport Runway Accident

Posted on:2011-04-26Degree:DoctorType:Dissertation
Country:ChinaCandidate:G M XuFull Text:PDF
GTID:1112330362458284Subject:Transportation planning and management
Abstract/Summary:PDF Full Text Request
Due to undertaking airplane take-off and landing, runway safety plays a very important role in airport safety as well as the whole air transportation system safety. Recently, with air transportation industry developed, more and more accidents are happening on airport runway or in the region that directly correlate with runway. Therefore, how to improve runway safety assurance ability, to identify, to estimate, to forecast and to control the risks existing in airport runway operations are the main problems concerned by the airport administering authority.In this thesis, two kinds of runway accidents–runway incursion and airplane hard landing are the study objects, and risk management theory is used to research on identification, estimation, forecasting and control of the risks existing in airport runway operations. The main content and innovation in this thesis may be summarized as follows:1. Based on the definition of runway accident risk and the depiction of risk property, runway accident risk management system is established, and the definition, goal, significance and procedure of risk idenfitication, risk estimation, accident forecasting and risk control are defined. Thus, the complete frame of runway accident risk management is developed, which provides the foundation of the later research is this thesis.2. Research on risk identification of runway accident. In view of airplane hard landing, the occurrence process is analyzed, and the major effect factors of airplane hard landing are determined through analysis of stress situation of airplane landing period. In view of runway incursion, runway incursion cases are analyzed, as well as questionnaire survey, the major effect factors of runway incursion are determined form the four aspects of human, correspondence, environment and management.3. Research on risk assessment of airplane hard landing. Based on research on the characteristic of airplane hard landing influencing factors, the method of airplane hard landing risk assessment based on hard landing intelligent diagnosis is proposed. Taking the values from flight quality monitoring database of airline as sample data, the diagnosis model based on Least Square Support Vector Machine (LS-SVM) is established, which expands diagnosis indexes from one index to several. Genetic algorithm is used to optimize the parameters of LS-SVM, thus the most superior diagnosis model is obtained to confirm the performance of the model. The example diagnosis results indicate that support vector machine model has the better diagnosis precision than neural network model. To improve the diagnosis accuracy, multi-kernel support vector machine is proposed, and it is applied to the example diagnosis of hard landing. The diagnosis results indicate that multi-kernel support vector machine surpasses the individual kernel support vector machine.4. Research on risk assessment of ruway incursion. Base on the design of risk index system, runway incursion risk is evaluated using extension evaluation method. In order to solve the problem in traditional correlation function, a new formula is put forward. Based on the classification of runway incursion, the risk estimation model based on human reliability analysis is established, and cognitive reliability and error analysis method is used to compute human failure probability.5. Research on forecasting of airplane hard landing and runway incursion. Based on the construction of training sample, the forecasting model of airplane hard landing and runway incursion based on Least Square Support Sector Regression (LS-SVR) is established. In view of the model parameter's optimization question, multi-layer adaptive grid search algorithm is proposed, which tuning the search step adaptively in every optimization region to optimize the model's parameters. The example results show that LS-SVR forecasting model is feasible and accurate for the trend prediction of airport runway accident.6. Research on risk control of runway accident. In view of safety manager, risk control method of runway accident based on risk defense is proposed. The risk defense system of runway accident is constructed. In view of the personnel whom runway accident involves, suggestions and measures are proposed for preventing runway accident.
Keywords/Search Tags:Airport safety, runway accident, risk management, support vector machine, extension evaluation, human reliability analysis
PDF Full Text Request
Related items