Modeling the Effect of Prolonged Wakefulness on Human Performance Using a Dual-Process QN-MHP-U

Accurately predicting human performance during prolonged wakefulness is essential for safety in high-risk environments, where operators are often required to maintain wakefulness over extended periods without sleep. The present study integrates the circadian process (Process C) and the sleep homeostatic process (Process S) into the Queuing Network-Model Human Processor-Unified (QN-MHP-U) via Server 9 to predict human performance during prolonged wakefulness. The model is applied to a system monitoring task, where operators must monitor analog gauges and warning lights and respond promptly to abnormal signals—a task type that is both representative of real-world monitoring duties and highly vulnerable to arousal decline. The model predicts reaction time on this task over 25 hr of wakefulness, achieving an RMSE of 0.41. The developed computational model can be further applied to domains requiring long-duration work, such as aviation and nuclear power operations.