Temporal Integration Framework for Segment-Level Cumulative Ergonomic Exposure Modeling

الملخّص

This paper proposes a vision-based temporal integration framework for segment-level cumulative ergonomic exposure modeling. Joint angles derived from OpenPose keypoints are mapped to REBA scores at six anatomical segments. A recovery-weighted exposure model quantifies cumulative risk by integrating posture severity, duration, and frequency while accounting for tissue recovery dynamics. Pattern recognition algorithms identify hazardous signatures without forecasting future postures. Validated through simulation of a manual handling task  (repetitive packaging) calibrated to empirically observed score distributions, the framework demonstrates that cumulative exposure metrics reveal critical risk patterns invisible to instantaneous scoring, specifically, sustained exposure to minimum-risk postures (REBA Score 2) generates substantial cumulative burden despite the absence of extreme scores. This pilot study establishes methodological feasibility for temporal integration in ergonomic assessment.

الكلمات المفتاحية

المؤلفون

1. 1ecole national polytechnique
2. 2Ecole National Polytechnique