The study shows that a deep learning model trained on survey responses from over 1,800 mining workers can predict sleep-related features that, when fed into the SAFTE model, generate fatigue readiness (ReadiScores) comparable to those derived from wearables.
This extensive research looked at biomathematical fatigue models applied to a variety of laboratory and field scenarios, such as railroad and aviation operations. The study concluded that fatigue models, such as the SAFTE Fatigue Model, are a reliable predictor of fatigue impairment and performance.
The SAFTE Fatigue Model uses a simple scoring system to measure fatigue impairment and cognitive effectiveness. It found that lower SAFTE Alertness Scores lead to an increase in the risk of occupational accidents.
It has been scientifically proven that workplace safety is compromised when people are fatigue impaired. This study validates the use of the SAFTE Fatigue Model in the aviation industry, as it relates to fatigue measurement and prediction, and its effect on human performance.
Sleep-tracking devices are growing in popularity and in recent studies have performed well against gold standard sleep measurement techniques. This study revealed Readi is a “promising alternative to research-grade actigraphy” under natural home sleep conditions.
The SAFTE Fatigue Model was developed and extensively studied under the extreme conditions encountered by subjects in the US military. The DoD has long recognized the effects of fatigue on physical and cognitive performance, and this paper further validates its ability to accurately measure and predict it.
The wrist-worn ReadiWatch (Readiband) is able to objectively determine human sleep and wake periods using the sophisticated technology of actigraphy. When compared to clinical polysomnography (PSG), the ReadiWatch (Readiband) is 93% accurate and far more practical for long-term, day-to-day sleep measurement.
A study by the Univ. of Illinois and St. Francis Medical Center used the SAFTE Fatigue Model and ReadiBand technology to measure resident doctor fatigue and effectiveness under real-time hospital conditions. The research revealed that shift scheduling had a detrimental effect on effectiveness scores.
Research done at Harvard Medical School on fatigue modeling, using the SAFTE Fatigue Model and the ReadiWatch (previously Readiband) wearable device, found that 48% of residents were fatigued and a further 27% were impaired. Results were worse for night-float staff. Overall, residents’ fatigue levels were predicted to increase the risk of medical error by 22%.
