A human-machine learning curve for stochastic assembly line balancing problems

The Assembly Line Balancing Problem (ALBP) represents one of the most explored research topics in manufacturing. However, only a few contributions have investigated the effect of the combined abilities of humans and machines in order to reach a balancing solution. It is well-recognized that human beings learn to perform assembly tasks over time, with the effect of reducing the time needed for unitary tasks. This implies a need to re-balance assembly lines periodically, in accordance with the increased level of human experience. However, given an assembly task that is partially performed by automatic equipment, it could be argued that some subtasks are not subject to learning effects. Breaking up assembly tasks into human and automatic subtasks represents the first step towards more sophisticated approaches for ALBP. In this paper, a learning curve is introduced that captures this disaggregation, which is then applied to a stochastic ALBP. Finally, a numerical example is proposed to show how this learning curve affects balancing solutions.