Minimizing Low Back Cumulative Loading during Design of Manual Material Handling Tasks: An Optimization Approach

Abstract

OCCUPATIONAL APPLICATIONS We present a practical method for minimizing low-back cumulative loading that leverages digital human modeling capabilities and optimization using an evolutionary algorithm. We demonstrate use of the method in a simulated lifting task. Our results show that this method is robust to different routines for calculating cumulative loading. The proposed method can aid ergonomics engineers in addressing a potential risk factor early in the design stage, even in the absence of an established threshold limit value, and it provides a time saving by eliminating the need to adjust workplace parameters across many design possibilities. TECHNICAL ABSTRACT Background Excessive exposure to low-back cumulative loading (LBCL) has been implicated as a risk factor for developing pain or injury during manual material handling (MMH) tasks. However, addressing LBCL during conceptual work design is challenging because of a lack of an established and widely accepted LBCL threshold limit value. We therefore formulate the design challenge using an optimization framework aided by digital human modeling (DHM). Methods We constructed a hypothetical MMH task requiring lifting, carrying, and placement of boxes into 16 storage locations. External loads were composed of four different mass categories handled 250 times, with four different relative handling frequencies. Resulting low back compressive force time series were integrated according to four suggested methods. Subsequently, we defined our objective function and constraints, and obtained a solution using an evolutionary algorithm. Results The percentage agreement between the four different relative handling frequencies and integration methods ranged between 89.5% and 100%. Kendall’s coefficient of concordance values ranged between 0.74 and 1.0, indicating good to perfect agreement among the solutions. Conclusion There is consensus is that minimizing LBCL exposure is beneficial, particularly during task design phases. Our results show that, irrespective of the theoretical background pertaining to LBCL quantification, the method proposed produces a robust and largely similar solution, at least for the MMH scenarios we simulated. Our proposed approach takes advantage of DHM capabilities to simulate diverse MMH scenarios and provides solution estimates at the conceptual design phase. The proposed method can be expanded using multi-objective optimizations schemes and additional constraints to provide a solution that addresses multiple injury and fatigue pathways.