Prediction of Physical Training Demand and Dynamic Course Adjustment Model for Flight Attendants Based on the Evolution of Civil Aviation Industry Standards

With the continuous evolution of civil aviation industry standards, flight attendant physical training requires dynamic adaptation to changing occupational demands. Aiming at lagging standard response, insufficient quantitative demand prediction, and rigid curriculum systems, this study constructs a four-dimensional quantitative index system for civil aviation standard evolution. An Improved Sparrow Search Algorithm–Random Forest (IMSSA-RF) prediction model is proposed, and a dynamic course adjustment mechanism linking standards, demand, and training is designed. Based on 495 panel data samples from 15 colleges and 30 airlines during 2014–2024, the IMSSA-RF model achieves 95.7% prediction accuracy, outperforming ARIMA, RF, and LSTM models. The dynamic curriculum adjustment improves the physical fitness pass rate by 37.2% and job adaptation by 41.5%. Emergency physical fitness and high-altitude safety are identified as the core driving factors. The model provides a systematic and quantitative scheme for precision training and adaptive curriculum design. It also has engineering relevance for aviation scenarios involving wearable monitoring, wireless emergency communication, and electromagnetic-compatible cabin safety systems.