Affiliations: [a] Department of Kinesiology, Faculty of Applied Health Sciences, University of Waterloo, Waterloo, Ontario, Canada, N2L 3G1. Tel.: +1 519 885 4567 x7080; Fax: +1 519 746 6776; E-mail: callagha@uwaterloo.ca | [b] Faculty of Human Kinetics, University of Windsor, Windsor, Ontario, Canada, N9B 3P4
Correspondence:
[*]
Corresponding author
Abstract: This study examined the error induced in estimating cumulative low back loading for exposure to dynamic manual materials handling tasks by using either static or quasi-dynamic biomechanical models when compared to a dynamic model. Ten male subjects performed three sagittal plane lifting tasks at three different lifting speeds and using three different hand loads. Digitized video recordings and measured hand forces were collected in order to calculate cumulative L4/L5 spinal loading (compression, moment, joint shear, and reaction shear) using rigid link and single muscle equivalent biomechanical models. Cumulative loading was calculated using three modeling approaches: static, quasi-dynamic, and dynamic. The calculation of cumulative loading using the dynamic model was set as the "gold standard" and error in the static and quasi-dynamic approaches was determined by comparison with the dynamic model. The use of a quasi-dynamic model resulted in an average error of −2.76% across all 10 subjects, 3 tasks, 3 lifting speeds and 3 masses. The static model had an average error of −12.55%. The error in both modeling approaches was significantly effected by the type of task performed, mass lifted, speed of lift, and model variable examined indicating that neither model produced consistent errors across the lifting parameters. The small errors associated with the quasi-dynamic model indicates that it holds promise as a method to reduce the amount of data required to estimate cumulative loading yet still preserve the dynamic loading exposure of a manual materials handling task.
