Study on Temperament Characteristics of Air Traffic Controllers Based on BP Neural Network

Abstract

The working characteristics of air traffic controllers are closely related to their temperament type. The selection of controllers, controller posts, and the composition of controller teams should all consider the adapta-bility of temperament. Therefore, it is crucial to study the distribution of temperament types among air traffic controllers to ensure the safety of control system operations and to prevent civil aviation accidents. In this study, we used the Eysenck Personality Questionnaire (EPQ) to conduct a questionnaire survey of air traffic controllers and compared the results to a national norm control group. Our analysis demonstrated that the tem-perament type of air traffic controllers differs significantly from that of ordinary people. We concluded that the most suitable temperament types for control work are polysanguine, mucinous, choleric, and depres-sive, in that order. Finally, we developed a fitting model based on a BP neural network to assess the suitability of individual controllers for spe-cific control positions. We used the questionnaire data to establish a da-ta set and selected the minimum root mean square error as the loss function for back propagation training. The simulation analysis con-firmed the effectiveness of the fitting model.