Manufacturing jobs have some of the highest injury rates of any sector, often due to the high physical and mental fatigue of employees. In an effort to improve workplaces, researchers have designed a system of wearable sensors that rely on machine learning to monitor workers for signs of physical strain and fatigue. They hope that their new devices will help prevent accidents and injuries.
The design is detailed in a study published in the October issue by a team at Northwestern University PNAS nexus. To measure fatigue and physical health, researchers developed an interconnected array of six wearable sensors placed across a wearer’s torso and arms. These were linked to two depth cameras to measure joint movements and an HD webcam to analyze movement intensity, repetition and reduced strength over time. Once turned on, these devices continuously monitored heart rate, skin temperature and movement patterns. But since there are no widely accepted, universal biomarker metrics for fatigue, researchers relied on the wearer’s self-reported perceived exertion levels on a scale of 0-10, which they then fed into a machine learning model. Once trained, this model was then used to predict a user’s fatigue levels in real time to provide a “more nuanced understanding of the person’s physical state” than previous studies, researchers said.
“The adoption of new technologies for real-time fatigue prediction has the potential to revolutionize manufacturing by optimizing work schedules and implementing adaptive work/rest cycles, [while also] addressing the problem of a lack of deterministic biomarkers,” the team wrote in their paper.
To test their system, the group recruited 43 participants between the ages of 18 and 56, then tasked them with simulating two difficult manufacturing jobs: wiring harnesses and laying composite panels. However, in these scenarios, the volunteers also wore weighted vests weighing as much as 20 kg to rapidly progress fatigue levels to those typically felt at the end of a shift. From there, researchers monitored the sensor measurements, as well as the machine learning program’s predictions. Researchers even took their sensors (without the extra weights) outside the laboratory environment and offered them to actual manufacturing industry workers at two factory locations, who noted that the system was easy to use and unobtrusive.
[Related: BMW puts humanoid robots in South Carolina factory.]
According to the results of the study, one of the immediate conclusions is the confirmation that ‘real meaningful feedback’ about fatigue requires that effort be considered as a ‘continuous variable’. Many existing methodologies classify an individual as fatigued or non-fatigued, they wrote, which is simply not useful enough when implementing preventative safety measures.
Furthermore, given that everyone’s body is different, the best mix of physical indicators of fatigue may vary from individual to individual, depending on factors such as age, gender and weight. However, some universal trends were still observed across the two production tasks. For example, one of the clearest signs of exertion can be found in the fatigue levels measured in a user’s non-dominant arm. Another indicator could be found by measuring the intensity of a worker’s ambulatory movements using his chest sensor. The sensor recordings for common signs of fatigue, such as increased heart rate, increased body temperature and perspiration, also helped model for exertion assessment.
Researchers hope that sensor systems similar to theirs will help provide more accurate and helpful monitors of real-time manual labor fatigue in factories. To help with this, they have made their method designs available online as open access material.
However, technology is only as useful as regulators allow it. Therefore, the team recognized that it is critical that manufacturing companies adhere to ethical and responsible standards when using these devices.
“While our overarching goal with this research is to ensure worker safety, reduce risk and empower operators through active feedback, we recognize the ethical and legal considerations associated with deploying such systems in real-world work environments,” the researchers wrote. “We are hopeful that continued technical advances, including our efforts in predicting physical fatigue in manufacturing environments, will inspire constructive discussions about deployment.”
