Data-driven ergonomic risk assessment of complex hand-intensive manufacturing processes

Area of Science:

• Ergonomics and Human Factors
• Occupational Health and Safety
• Manufacturing Engineering

Background:

• Hand-intensive manufacturing tasks pose risks for musculoskeletal disorders.
• Existing ergonomic assessments like RULA and HAL lack granularity for hand and finger activity.
• Need for advanced risk assessment systems in manufacturing.

Purpose of the Study:

• Develop a data-driven ergonomic risk assessment system for hand and finger activity.
• Introduce a new, more granular ergonomic score for hand and finger risks.
• Enable better identification and mitigation of workplace risks in manufacturing.

Main Methods:

• Integrated a multi-modal sensor testbed to capture upper body pose, hand pose, and applied force.
• Collected data during hand-intensive composite layup tasks.
• Trained machine learning models to predict existing ergonomic metrics (RULA, HAL).

Main Results:

• Introduced the Biometric Assessment of Complete Hand (BACH) ergonomic score.
• BACH score offers greater granularity than RULA and HAL.
• Machine learning models effectively predicted RULA and HAL metrics for new participants.

Conclusions:

• The developed system provides enhanced ergonomic interpretability for manufacturing processes.
• Enables targeted workplace optimizations and posture corrections.
• Aims to improve worker safety and reduce musculoskeletal disorders.