Maximizing comfort in Assembly Lines with temporal, spatial and ergonomic attributes

Abstract

We aim at maximizing the comfort of operators in mixed-model assembly lines. To achieve this goal, we evaluate two assembly line balancing models: the first that minimizes the maximum ergonomic risk and the second one that minimizes the average absolute deviations of ergonomic risk. Through a case study we compare the results of the two models by two different resolution procedures: the Mixed Integer Linear Programming (MILP) and Greedy Randomized Adaptive Search Procedures (GRASP). Although linear programming offers best solution, the results given by GRASPs are competitive.

A case study at the Nissan Barcelona factory to minimize the ergonomic risk and its standard deviation in a mixed-model assembly line

“…(i) Minimizing both objectives simultaneously (ii) Subordinating one objective to the other one (iii) Solving the problem mono-objectively and determining the other objective afterwards In accordance with the second way, and taking the previous work (Bautista et al, 2016a) as a reference, a mathematical model to minimize the maximum ergonomic risk of the line first, and then, the ergonomic risk dispersion between workstations, is presented. Specifically, in this work, the ergonomic risk dispersion is measured through the standard deviation, unlike Bautista et al (2016a), where the average absolute deviation was considered. The parameters, variables, and the mathematical model formulation are shown below:…”

“…To define the neighborhood of a solution, and how to explore it 3. To define the stopping criterion based on runtime or number of iterations Specifically, the GRASP proposed in this paper is similar to that in Bautista et al (2016a). However, in this work, the main goal is to minimize the ergonomic risk of the critical workstation (station with greatest risk), and subject to this first objective, the second goal is to minimize the standard deviation (SD) from the ergonomic risks of the assembly line.…”

“…The results given by the proposed GRASP are compared with those obtained by two mono-objective mathematical models because MILP does not support hierarchical objective functions. Specifically, the exact procedures are the MILP-1 and MILP-2, wherein their objectives are minimizing the maximum ergonomic risk of the assembly line and minimizing the risk dispersion between workstations, respectively (see Bautista et al, 2016a).…”

“…_ model (see Bautista et al, 2016a): a. Its objective function forces an equal sharing of risk between all workstations by minimizing the average absolute deviations from risks of workstations and without considering the maximum risk minimization.…”

“…Later, a new mathematical model was proposed in order to reduce differences between workstations. Specifically, the new model minimized the average absolute deviation of the ergonomic risks of the set of workstations (see Bautista et al, 2016a).…”