A dynamic decision-making simulation model for pilots considering risk preference heterogeneity

With the increasing complexity of aviation safety decision-making, understanding pilots’ heterogeneous risk preferences and their influence on dynamic decision processes has become crucial. However, most studies largely rely on traditional models and lack rigorous mathematical predictive frameworks. To achieve this goal, a comprehensive risk preference classification model for pilots is proposed by integrating subjective scales with objective experimental data, categorizing pilots as risk-averse, risk-neutral, or risk-seeking. Subsequently, a dynamic decision-making framework grounded in the Extended Decision Field Theory (EDFT), interval-valued intuitionistic fuzzy number (IVIFN), and the OODA loop (Observe, Orient, Decide, Act) is developed to capture pilots’ cognitive and behavioral dynamics during task execution. Data from 30 participants, including scale assessments, behavioral measures, and EEG recordings, are used to validate the Pilot Risk Attitude Scale, classify pilots into three risk-preference categories, and extract EDFT parameters. Building on this foundation, a MATLAB-based simulation algorithm is then developed to model pilots’ cognitive and behavioral dynamics across different risk-preference categories during landing alternatives. Results show close agreement between simulated and experimental outcomes (Pearson’s r = 0.981), confirming the model’s predictive power. Furthermore, the model effectively discriminates dynamic decision-making behaviors among pilots with differing risk preferences. Meanwhile, extended decision-making intervals are found to enhance the selection of preferred alternatives, whereas elevated decision thresholds produce delays in choosing favored options.