You see it in the news: worker fatigue leading to accidents, with an employee hurt or even killed, or causing injury to others. But that’s not the only possible outcome of worker fatigue. Chronic injuries from repetitive movements, reduced work output and reduced work quality can also result from employees becoming tired on the job.
Zahra Sedighi Maman, PhD, was involved in a research project using data analytics to predict the occurrence of physical fatigue in manufacturing environments and to monitor its development over time, with the goal of preventing worker injuries.
“Worker fatigue is an important issue. We need to detect and manage it,” said Dr. Sedighi Maman, assistant professor of decision sciences and marketing. “A recent survey has shown that around 58 percent of manufacturing employees are fatigued at work.”
She points out that more than 12 million people are employed in manufacturing in the United States alone. Many more millions work in factories abroad.
Dr. Sedighi Maman and fellow researchers used small, nonintrusive wearable sensors on eight volunteers doing simulated manufacturing tasks for three hours at a time. Sensors attached at five locations (heart, wrist, hip, ankle, torso) on the body collected an extensive amount of data on numerous aspects of the participants’ motion throughout these sessions in an effort to detect signs of fatigue, such as slowing down, changes in posture, tremors, jerky movements and more.
Using the data, the researchers developed a model for predicting, detecting and monitoring the fatigue that results from employees performing typical manufacturing tasks at their daily jobs. This model is also widely adaptable to other activities, said Dr. Sedighi Maman. The researchers further discovered that changes in wrist and hip movement may be more useful for detecting physical fatigue than changes in movements at other sensor locations.
Their paper, “A data-driven approach to modeling physical fatigue in the workplace using wearable sensors,” can be accessed online at Applied Ergonomics.
“Companies are interested in wearable sensors to monitor the development of fatigue in employees, and we believe our research will encourage management to invest in data-driven monitoring to prevent occupational injury,” said Dr. Sedighi Maman. In fact, according to a survey conducted among U.S. manufacturing safety professionals, 54 percent are interested in using wearable technology for this purpose, she added.
In another paper, now under review for publication, she proposes a unified means of managing physical fatigue in the workplace through fatigue detection, identification, diagnosis and recovery, with the use of wearable sensors.
“It’s possible to use wearable sensors to predict fatigue, but we also have to think about interventions,” she said. “Can we come up with a model for managing fatigue from the beginning using interventions? We worked to design a proper framework of analytics for any physically fatiguing task, in manufacturing or other industries.”
Dr. Sedighi Maman’s new research also discovered that fatigue could be detected using only one sensor. This is an important finding for manufacturers. “One sensor costs much less than four or five, with no loss in fatigue prediction,” she explains.
Her research specialties include data analytics, data mining and data visualization, with an overall focus on ergonomics and safety and on healthcare. The overall goal of her research, she said, is “to make a better life for people.” She holds a PhD in industrial and systems engineering.