An Integrated Lifting Predictive Model for Lumbar Injury Risk Assessment

Manual lifting is a common activity in the industry. For reducing the lumbar injury risk during lifting, predictive injury risk assessment models have been widely used. However, limited attention has been paid to the dynamic lifting prediction in different lifting techniques, resulting in limited application of existing models. This paper proposed an integrated model, in which dynamic lifting motions were generated by an optimization-based model, and the lumbar joint reaction forces were then estimated by a musculoskeletal model. This model was validated and then applied to explore the effect of lifting techniques (squat and stoop lifting) on the lumbar load under different lifting conditions. Results show that the proposed model leads to accurate compressive forces and the risk level on L5S1. The most commonly advised squat lifting is proven to be safer when the box is light and close enough to the person, but the conclusion may be reversed when the box is further away.