Presentation
Investigating 24- to 48-hour Forecast of Offshore Worker Alertness and Vigilance using Multimodal Sources: A Proactive Fatigue Monitoring Paradigm
DescriptionThe oil and gas industry has faced significant safety risks, often stemming from prolonged physical and mental exertion or sleep deprivation. While fatigue forecasting (predicting future fatigue states) has been explored in controlled lab settings for brief periods, there is a notable gap in predicting fatigue levels over multiple days, which can be crucial in high-risk occupations like oil and gas. Current fatigue assessments in the oil and gas industry rely heavily on subjective questionnaires, which may have less correlation with objective metrics. Various approaches, such as combining sleep quality and performance metrics, have been investigated to establish reliable fatigue markers. However, there's a need to focus on forecasting fatigue states to enhance mitigation strategies. This study aims to address these gaps by examining different forecasting algorithms, features, and time horizons using the Psychomotor Vigilance Task (PVT) reaction times to improve fatigue prediction in high-risk industries.
Contributors
Event Type
Lecture
TimeWednesday, September 11th10:25am - 10:45am MST
LocationFLW Salon H
Augmented Cognition
Extended Reality