The three-process model of alertness and its extension to performance, sleep latency, and sleep length

This paper summarizes a mathematical/computer model for predicting alterness/performance in daily living. The model uses sleep data as input and contains a circadian and a homeostatic component (amount of prior wake and amount of prior sleep), which are summed to yield predicted alertness (on a scale between 1 and 16) as well as performance on monotonous tasks. The model includes an identification of levels at which the risk of performance/alertness impairment starts, as well as prediction of sleep latency and time of awakening of sleep episodes. It is suggested that the model may be used to evaluate work/rest schedules in terms of sleep-related safety risks.