Research Of Algorithm Of Training Effectiveness On General Trainer

At low altitude with the gradual implementation of the open policy, China will for a unprecedented development opportunity in general aviation. With pilots shortage,how to choose the trainer to improve training effectiveness to reduce training costs, is one of the pressing concerns problem to flight training schools. Training effectiveness evaluation can provide an effective basis for the selection of trainer, allowing trainers to choose a more targeted.the Masefield-Burdak training effectiveness evaluation model is introduced and the process of general trainer training effectiveness evaluation process were analyzed in this paper. For the indicators standardized and indicators weight quantitative problem,A new effective algorithm is put forward based on more indicators integrated evaluation theory, which evaluation indicators is divided into benefits type, and cost type and interval type; and respectively using linear transform method, vector transform method, very poor transform method, and share transform method, and index transform method, method on indicators for specification. To get e evaluation indicators matrix, compatible degrees is defined to select optimal standardized method based on Spearman related coefficient. The weight matrix to evaluation index to flight discipline is calculated by experts subjective score and objective assigned right method.In order to avoid shortage of evaluation theory, An improved model was proposed.The cross efficiency factor is introduced to modify the trainer effectiveness evaluation model above based on DEA cross efficiency model. About input and output indicators of cross-efficiency model, efficiency indicators is presented as output indicators for cross efficiency model for a evaluation unit, cost index as input index. Interval indicator you can use the corresponding data transformation into efficiency indicators, it can also be used as output index.To verify the actual of The improved method, practical example is used respectively, and the results is accord with the actual, not only demonstrates theimproved training algorithm to evaluate the effectiveness of the results is objective can have accuracy improved.