Occupational Safety within Non-Routine Manufacturing Processes: Evaluating the Validity of Task-Based Ergonomic Assessments

2025-07-27

AI Summary

This study addresses the insufficient construct validity of task-oriented human factors ergonomics assessment methods in non-routine manufacturing, specifically examining the risk of systematic underestimation of musculoskeletal hazards under conditions lacking direct physiological measurement. Innovatively integrating a Multitrait–Multimethod (MTMM) matrix with video-based content analysis—and using inertial motion capture data and the Borg Rating of Perceived Exertion (RPE) scale as criterion measures—it provides the first systematic validity evaluation of such assessments. Results reveal weak convergent validity (r = 0.149–0.243), and although discriminant validity is statistically significant (p < 0.001), its practical discriminative power is limited—leading to systemic underestimation of ergonomic risk and misclassification of risk severity levels. Three critical determinants of validity impairment are identified, and a technical pathway for improving cumulative load modeling is proposed, thereby strengthening the methodological foundation for evidence-based human factors interventions in complex manufacturing environments.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Intelligent Robots: State Estimation
• Machine Learning: Matrix & Tensor Methods

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Direct measurement ergonomic assessment is reshaping occupational safety by facilitating highly reliable risk estimation. Industry 5.0, advocating human-centricity, has catalysed increasing adoption of direct measurement tools in manufacturing industries. However, due to technical and feasibility constraints in their practical implementations, especially within non routine manufacturing processes, task based approach to ergonomic assessment is utilized. Despite enabling operationalization of robust ergonomic assessment technologies within complicated industrial processes, task based approach raises several validity concerns. Hence, to ascertain functional utility of the resultant safety interventions, this study evaluates the construct validity of task based ergonomic assessment within non routine work utilizing Multitrait multimethod (MTMM) matrix followed by video-based content analysis. Ergonomic exposure traits were collected for 46 participants through direct measurement and self reported techniques utilizing inertial motion capture and Borg's RPE rating scale respectively. Findings include unsubstantiated convergent validity (low same trait correlations from 0.149 to 0.243) and weak evidence of discriminant validity with statistical significance (p value less than 0.001). The study also identifies three primary factors undermining construct validity through video based content analysis. Findings also elucidate misinterpretation of ergonomic risk and action levels. Therefore, practical implications entail underestimation of actual ergonomic risks when estimated through task based assessment. This highlights the need for enhancement in ergonomic assessment technologies focused on cumulative load analysis compatible within diverse industrial processes.

Problem

• Evaluating validity of task-based ergonomic assessments in non-routine manufacturing
• Assessing construct validity using MTMM matrix and video analysis
• Identifying factors causing underestimation of ergonomic risks in task-based methods

Innovation

• Direct measurement tools enhance ergonomic risk estimation
• Task-based approach used for non-routine processes
• MTMM matrix validates ergonomic assessment methods