Study On The Index System Of The Airport Passengers Classification Based On Risk

In recent years,with the development of civil aviation industry,the passenger throughput is increasing,and the passenger structure is becoming more and more diverse,which causes the long waiting time and the low quality of security.The great pressure has been on security inspection work by the diversification of passenger structure and the increase of passenger throughput.Traditional security mode has gradually failed to satisfied the needs of airports and passengers.Therefore,the main problem is to reduce the passenger waiting time,improve the quality of security inspection,minimize the cost of service resource and maximize the utilization of security resource.Firstly,in order to construct the index system of airport passengers classification based on risk,the meanings of passenger profiling were studied,and a definition of passenger classification security system was proposed in this part.Based on seven foreign airport passenger classification security systems of CAPPS I,CAPPS II,Secure Flight,Percheck,Smart Security,Trusted Travelers and the Privium were introduced.However,because of the lack of the related information about index system,personal credit evaluation index system of banks and sesame credit were used for reference.Secondly,based on construction principles of index system and the related index systems,the index system of airport passenger classification based on risk was established preliminarily,and meanings of indexes were introduced in this part.Then,according to initial index system,a questionnaire was designed to investigate,and the reliability and validity were analyzed by SPSS22.0.After analyzing and screening,47 indexes were finally determined,including 5 primary indexes,natural condition,occupation status,economic condition,credit situation and flight condition.So a optimized index system of the airport passengers classification based on risk was finished.Finally,the levels of passenger classification were analyzed and defined.According to the study and comparison of 4 classifiers,BP neural network classifier was selected for passenger classification.In view of limitations of BP neural network,PSO algorithm was presented to optimize BP neural network classifier.After the quantification of riskclassification indexes,MATLAB was programmed to implement the function of classification.And 5 sets of data were used to ascertained the risk levels of passengers by the PSO-BP neural network classifier.