Study On The Early Warning Model For Ground Service Personnel Human Error Of Airport Airfield

Safety is the lifeline of air transport, only to ensure safety, can the high efficiency of airbe converted to yield. The development of modern engineering techniques greatlyincrease the safety and reliability of the equipment, and the plane flight accident ratecaused by equipment failure has dropped from80%at the beginning of the20thcentury to3%today. Related research at home and abroad show that human error hasbecome the main factors affecting air safety.Airport airfield security is an important part of the aviation safety, and is not justabout whether air transport can function normally, also directly affects the safety ofpassengers, cargo and aircraft. Compared with the air traffic, the airport securityincidents caused by airport airfield ground service errors are often less eye-catching,but it occurs not only bring casualties, aircraft and equipment damage and other directloss, but also may result in delays and serious air traffic accidents and other indirectloss. Because the work content is monotonous and repetitive, and time pressure isbigger, the ground service personnel are often prone to human error. So the groundservice personnel human error risk early warning research is of great significance toimprove the level of safety management of airport flight zone.Based on related research at home and abroad, this thesis summarizes andanalyzes the connotation, characteristics and classification of human error in groundservice personnel, and expounds the theory model. This thesis analyses the causes ofthe ground service personnel human error on the basis of the systemic human errorreduction (SHERPA) and prediction approach hierarchical task analysis (HTA), putsforward the model of human error mechanism, establishes the index system, andscreens indexes by analytic hierarchy process interval estimation method. Establishthe human error risk early warning model on the basis of early warning indicators, bythe use of questionnaire survey and Noisy or gate model to determine the Bayesiannetwork prior information. For getting parameters of Bayesian network, BNT wasadopted, which is based on the MATLAB. Conducted a predictive reasoning and diagnostic reasoning based on Monte Carlo algorithm, and get service personnel forthe human reliability level and the key factors that lead to human errors the keyfactors that lead to human errors. Finally, proposes the countermeasures andsuggestions to deal with human error based on the simulation analysis conclusions. Asensitivity analysis of human error early-warning indicators was performed by usingattenuation of entropy. On this basis, put forward to forecasting the trend of the keyearly warning indicators of ground service personnel human errors by using themarkov chain prediction model, and provide the basis for the daily safetymanagement to take preventive measures.