NLP-based ergonomics MSD risk root cause analysis and risk controls recommendation.

Area of Science:

• Occupational Health and Safety
• Artificial Intelligence in Ergonomics
• Human Factors Engineering

Background:

• Musculoskeletal disorders (MSDs) are a significant workplace concern, necessitating effective ergonomics assessments.
• Artificial Intelligence (AI) offers potential for more accurate and efficient ergonomics evaluations.
• Current AI approaches often lack actionable guidance for risk reduction, focusing solely on risk scores.

Purpose of the Study:

• To develop a holistic job improvement process for reducing MSD risk.
• To automate root cause analysis and provide specific control recommendations for MSDs.
• To integrate Natural Language Processing (NLP) and Machine Learning (ML) for enhanced ergonomics assessments.

Main Methods:

• Utilizing deep learning-based NLP (Part of Speech tagging, dependency parsing) on job action descriptions.
• Applying an expert-based Machine Learning (ML) system for root cause identification of MSD risks.
• Linking identified root causes to recommended control strategies for risk mitigation.

Main Results:

• The proposed framework successfully performs automatic root cause analysis of MSD risks.
• Action-object inferences from NLP guide the ML system to pinpoint specific work-related causes.
• The system recommends targeted control strategies, such as modifying equipment (e.g., caster size), to reduce MSD risk.

Conclusions:

• This AI-powered framework provides a comprehensive approach to ergonomics, moving beyond risk scoring to actionable job improvement.
• The integration of NLP and ML enables efficient identification of MSD causes and effective risk reduction strategies.
• The proposed method enhances the ergonomics assessment process, leading to more efficient and effective prevention of musculoskeletal disorders.