Predicting Fatigue with Machine Learning

Article by Paul Adair, WDC Staff Writer, published for Women Driving Change.

A lack of quality sleep, performing monotonous tasks, or driving for long periods without a break can cause fatigue. Fatigue is like alcohol impairment and includes symptoms such as decreased alertness and impaired decision making, or even nodding off entirely. This is obviously concerning for anyone who works in a 24-hour industry, like transportation, as well as for those who share the road with professional drivers.

One approach used to tackle the problem has been using biomathematical modeling to predict and prevent fatigue-induced risks, such as the Sleep, Activity, Fatigue, and Task Effectiveness (SAFTE) model.

But this model first needs data to be truly effective.

Fatigue Science is the global leader in predictive fatigue management information systems for heavy industry, military, and transportation. Its ReadiWatch wearable device collects invaluable sleep information via actigraphy, or high frequency accelerometry, that analyzes the movement of the wrist, which is then input into the SAFTE model.

However, delivering predictive fatigue intelligence without the use of wearables has been among the biggest asks from Fatigue Science clients. Answering the call, Fatigue Science has launched its industry leading Readi FMIS, which uses machine learning and data science to allow customers to receive personalized fatigue predictions without having to rely on a wearable.

“With machine learning, we can now bring the safety- and productivity-enhancing benefits of predictive fatigue technology to a much wider array of customers who are not yet ready to adopt wearables,” said Fatigue Science President & CEO, Andrew Morden in a press release.

Read the full article on womendrivingchangemagazine.ca