Research On Competency-Oriented The Evaluative Method Of Screening Test For Flight Students

In recent years,the strong demand for pilots in the civil aviation industry has led to a sharp expansion in the number and scale of domestic flight training schools.At the same time,the frequent occurrence of unsafe flight events due to crew reasons in current civil aviation transportation production reflects the shortcomings of existing flight training links and the urgent need to improve flight training management,highlighting the importance of pilots’ core competencies.Initial flight training is an important stage in the development of pilots’ core competencies,aimed at helping students lay the foundation of flight technology required for future airline transportation flights.However,the existing flight training evaluation has limitations such as the lack of comprehensiveness of evaluation indicators,the lack of scientific design of evaluation standards,the lack of objectivity and stability of evaluation methods,and the failure to effectively utilize the value of flight training data,which cannot meet the requirements of training ideas in the new era.Although the International Civil Aviation Organization(ICAO)has proposed an assessment framework for the core competencies of airline transport pilots,there is a lack of operational quantitative rating methods for assessing various competencies.Therefore,it is particularly urgent to reform teaching and training concepts,models,and evaluation methods.Especially as a large number of newly registered flight training schools enter the market and the training scale continues to expand,a large number of new flight instructors and inspectors quickly become the main force of flight teaching.They still need to accumulate experience in understanding and grasping the requirements of the training program,especially in ensuring training safety and preventing flight overruns.To this end,the Civil Aviation Administration clearly proposes to deepen the reform of flight training,"build a strongly supported,collaborative,efficient,open and innovative flight training system with Chinese characteristics for the new era,and provide an important strategic fulcrum for comprehensively building a strong civil aviation power.".Based on the new era training idea of "implementing flight training based on core competencies",this thesis constructs a competency oriented evaluation model for flight training using the evaluation dimensions in the VENN criteria and subject and observation points as evaluation indicators,taking the screening,inspection,and evaluation methods of flight cadets as the research object,and taking competency as the guidance.This article focuses on the following aspects:First,accurately understand core competency requirements.Systematically analyze documents such as flight training practice examination standards and curriculum syllabuses issued by the Civil Aviation Administration,comprehensively comb the training requirements for students’ technical and non-technical abilities at various stages of initial flight training using expert survey methods,focusing on selecting courses,and analyzing the behavioral characteristics of various abilities.Referring to ICAO’s criteria for determining the core competencies of airline transport pilots,the core competencies of the selected course were summarized and refined.The results showed that the competency characteristics of the course were mainly distributed in "FPM Flight Path Management Manual Flight",and were refined to the five observation points corresponding to the competency,"OB4.1-OB4.5".Secondly,the construction of core competency evaluation model.Firstly,establish a core competency evaluation index system.The expert survey method is used to initially select core competency evaluation indicators,and then the construction and statistical analysis of correlation matrices are used to screen out evaluation indicators that can represent the core competency required for the selected courses and subjects.Based on the idea of analytic hierarchy process,a multi-dimensional and multi-level competency evaluation indicator system of "curriculum competency → curriculum competency indicators → subject competency indicators → observation points" is constructed.The second is to design a core competency measurement scheme.Using the expert survey method,with the help of the rich evaluation experience of flight experts and traditional work orders,the rating criteria for each competency item and its observation points are defined.Then,according to the VENN criterion,the evaluation indicators under the "How often" and "How many" dimensions are statistically analyzed using training big data.Based on the constructed competency rating model,the observation points of each competency indicator are scored,and the ratings for each level of competency item are obtained in turn.Third,the validation of the evaluation model.Based on three machine learning algorithms,namely,CART decision tree,support vector machine,and BP neural network,the feasibility and reliability of the evaluation scheme were verified,and the accuracy was higher than 80%,fully demonstrating the feasibility and reliability of the designed competency rating model.The effectiveness of the evaluation scheme is verified based on the correlation analysis of Persons.By analyzing the correlation between competency rating statistical data and the rating results provided by examiners,it is shown that there is a significant correlation between the two,confirming the effectiveness of the competency based evaluation scheme.The research results of this article contribute to improving the comprehensiveness,objectivity,and stability of initial flight training evaluation in civil aviation,and lay the foundation for deepening the reform of flight training in the civil aviation industry and implementing the full life cycle management of pilot technical capabilities.At the same time,a set of effective and skill training oriented big data analysis methods will be formed,which will also have important reference value for the analysis and application of training data,learning performance analysis and evaluation of other courses in the initial flight training stage.The formed flight training data samples can also provide data support in the field of flight training for big data,artificial intelligence,and management science research.