Physiological Assessment and Illustrative Cyber–Physical Modeling of Operator-in-the-Loop Railway Control Dynamics Under Adverse Operational Scenarios

Connecting train operators’ physiological responses with activation delay consequences remains a methodological challenge in railway safety research. This study combines physiological characterization with an independent illustrative cyber–physical human-in-the-loop control and braking model. Eight train operators, divided equally into experienced and non-experienced groups, completed resting, normal-weather, severe-weather, and car-crash scenarios. Group-level responses were characterized descriptively using scenario contrasts and joint principal component analysis (PCA), with pairwise analysis of variance (ANOVA), multiplicity adjustment, and autoregressive generalized least squares retained as model-conditional sensitivity summaries. A modified proportional–integral–derivative (PID) formulation incorporated prescribed gains, phase margins, and activation delays. From rest to the car-crash phase, heart rate increased by 5.21% and 8.98%, and respiratory rate by 5.14% and 9.23%, in the experienced and non-experienced groups, respectively. Root mean square of successive differences (RMSSD) and standard deviation of normal-to-normal intervals (SDNN) decreased in both groups, with lower values in the experienced group. Interpretation concerns the observed group profiles and scenario sequence. Joint PCA captured 83.82% of variance (61.40% and 22.42%). At 120 km/h and 3.5 m/s2 deceleration, increasing activation delay from 0.5 s to 1.8 s extended stopping distance by 43.33 m (175.40–218.73 m). Physiological profiles and multivariate characterization identify patterns for subsequent validation, while the independent control–braking calculations quantify the consequences of modeled response dynamics for operator monitoring and adaptive assistance.