August 2026 Volume 8

SAFETY

movement precision and compensate for declining physical performance. Cognitive fatigue can also reduce concentration and motor planning, making physically demanding tasks feel more strenuous and increasing the likelihood of movement errors 7,12 . In high-risk environments such as forging, where workers are continually handling hot materials, heavy tools, and high-impact equipment, the combined effects of physical and decision fatigue may substantially increase the risk of accidents, reduced product quality, and musculoskeletal injury. Overall, fatigue mechanisms identified in other repetitive occupations provide a useful framework for understanding fatigue in forge workers. Although forging generally involves considerably higher mechanical loading, the underlying physiological responses, including reduced neuromuscular performance, impaired movement accuracy, diminished joint stability, and declining cognitive performance, are expected to follow similar trends. Recognizing both the physical and cognitive components of fatigue can improve risk assessments and support the development of more effective ergonomic interventions. Use of Exoskeletons One intervention that has been frequently considered is the use of wearable occupational exoskeletons (Figure 2). Upper limb exoskeletons are designed to partially support the weight of the arms and tools, reducing the muscular effort required during repetitive or sustained tasks. By decreasing shoulder and upper-arm loading, these devices may delay the onset of muscular fatigue, reduce compensatory muscle recruitment, and lessen cumulative joint loading throughout a work shift. Reduced physical effort may also indirectly lessen cognitive workload by allowing workers to devote fewer mental resources to maintain posture and movement accuracy. Although additional research is needed to assess what type of exoskeleton would be best suited for a forging environment, occupational exoskeletons represent a promising strategy for reducing fatigue, improving endurance, increasing productivity, and lowering the long-term risk of musculoskeletal injury.

of relying only on questionnaires or visual observations, AI models can analyze data collected from wearable sensors, such as electromyography (EMG), electrocardiography (ECG), inertial measurement units (IMUs), and heart rate variability, to estimate fatigue continuously during work 13 . AI can also analyze video recordings using computer vision techniques. From video, the system can detect changes in posture, upper-limb movement, joint angles, movement speed, hammering rhythm, and tool trajectory that may indicate the development of fatigue 14 . Video-based monitoring is attractive because it is non-contact and does not require workers to wear additional sensors. However, combining video with physiological signals generally provides more reliable fatigue assessment because it captures both movement changes and the body's physiological response to fatigue 13,14 . Recent studies have also explored anomaly detection, in which the model first learns a worker's normal movement and physiological patterns and then identifies deviations from this baseline that may indicate fatigue. This approach is well suited for industrial environments because it does not require large amounts of labeled fatigue data and can adapt to differences between individual workers 15 . Recent deep learning methods have further improved fatigue prediction by combining information from multiple sensors and video, allowing more accurate and personalized fatigue monitoring 13,16 . Future research may also benefit from physics-informed AI models that combine biomechanical knowledge with multimodal sensor data, which could improve both the accuracy and interpretability of fatigue assessment in occupational settings 17 . For forge workers, these methods could provide continuous monitoring of fatigue, identify workers at greater risk of injury, and offer an objective way to evaluate interventions, such as occupational exoskeletons, by measuring changes in fatigue throughout a work shift. Future Directions There is an undeniable need to both (i) assess the extent of musculoskeletal pains that workers in the forging industry experience, and (ii) develop solutions that minimize the likelihood of these impacting the health of the worker. Wearable sensors, remote monitoring, technology for offloading the upper extremity and the use of AI are key ingredients in any path towards creating environments where musculoskeletal fatigue does not compromise safety and/or the production output of the manufacturing facility. References: 1. Reville, Robert T., Jayanta Bhattacharya, and Lauren R. Sager Weinstein. 2001. "New Methods and Data Sources for Measuring the Economic Consequences of Workplace Injuries." American Journal of Industrial Medicine 40(4): 452–463. 2. Chaffin, Don, Lawarence J. Fin. 1991. A National Strategy for Occupational Musculoskeletal Injuries — Implementation Issues and Research Needs. Centers for Disease Control and Prevention. DHHS (NIOSH) Publication No. 93-101, 1-27. 3. Nelson NA, Hughes RE. Quantifying relationships between selected work-related risk factors and back pain: a systematic review of objective biomechanical measures and cost-related health outcomes. Int J Ind Ergon. 2009 Jan 1;39(1):202 210. doi: 10.1016/j.ergon.2008.06.003. PMID: 20047008; PMCID: PMC2662685. 4. You, Heecheon & Simmons, Zachary & Freivalds, Andris & Kothari, Milind & Naidu, Sanjiv & Young, Ronda. (2004). The development of risk assessment models for carpal tunnel syndrome: A case-referent study. Ergonomics. 47. 688-709. 10.1080/0014013042000193291.

Figure 2: Example of an exoskeleton being used to offset the weight of a grinder. In this case, the exoskeleton could reduce fatigue in shoulder muscles.

Role of Artificial Intelligence Artificial intelligence (AI) is becoming a useful tool for measuring fatigue in physically demanding workplaces. Instead

28 FIA MAGAZINE | AUGUST 2026