Many assembly line balancing algorithms consider only the precedence relations between the tasks and task duration while attempting to minimize cycle time. These approaches fail to account for the physical demands placed on the personnel who operate these assembly lines. Such physical demands can include work requiring repetitive and forceful manual exertions. Disregarding these human physical requirements when balancing assembly lines may contribute to the development of work related musculoskeletal disorders in the work force. Three line balancing heuristics, which incorporate ergonomic design criteria, were developed: a multiple ranking heuristic, a combinatorial genetic algorithm, and a problem-space genetic algorithm. Each algorithm was designed to minimize both assembly line cycle time and the loss grip strength incurred across workstations due to fatigue. The most fit solutions found by each heuristic were evaluated and compared in terms of cycle time and fatigue (i.e. maximum loss in grip strength). Each heuristic searched for solutions under both I-shaped and U-shaped assembly line configurations using 100 line balancing problems. Results of the statistical analysis strongly suggest that the evolutionary based approaches and U-shaped assembly line configurations allow tor the formation o f assembly line balances which minimize cycle time and the physical workload placed on line participants. Assembly line modifications based upon the results of this research could prove beneficial to industry from both a production and ergonomics perspective.
Keywords:
Assembly Line Balancing; Ergonomics; Evolutionary Computation; Genetic Algorithms; Grip Strength; WRMSD