Fatigue risk management tools must be based on solid science and deliver accurate assessments. A model that is not sufficiently sensitive to the factors that lead to performance error and accidents can point to costly and unnecessary changes on the one hand, or provide a false sense of security by not detecting genuine fatigue risks when they exist.
The SAFTE® model and the SAFTE-FAST system have been subjected to independent validation showing that they can predict variations in laboratory performance with less error than any other available model. Key operational studies demonstrate that SAFTE predicts elevations in railroad accident risk based on an analysis of work schedule information alone.
A related study showed that the SAFTE model predicts the elevated cost of fatigue related accidents. Employees expected to be the most fatigued were about 70% more likely to have an accident and accidents that were associated with low levels of predicted performance were, on average, four times more expensive than accidents caused by rested employees.
A study of fatigue in cabin crew showed that the SAFTE model could predict changes in the speed of performance on a vigilance test (PVT) based on actigraphy-based sleep data while another study of cockpit crew showed that the model can predict patterns of sleep for individual pilots with a minute-by-minute accuracy of 85% with overall average sleep amounts predicted to within a minute of actual amounts.
Analysis of the Relationship between Operator Effectiveness Measures and Economic Impacts of Rail Accidents
Validation and Calibration of a Fatigue Assessment Tool for Railroad Work Schedules, Summary Report
Flight Attendant Work/Rest Patterns, Alertness, and Performance Assessment: Field Validation of Biomathematical Fatigue Modeling
Fatigue Models for Applied Research in Warfighting