Modeling and solving multi-objective mixed-model assembly line balancing and worker assignment problem

• This paper deals with mixed-model assembly line balancing and worker assignment problem.
• Two objectives including cycle time and total operating costs related to workers are considered as performance criteria.
• Goal programming approach is used to deal with bi-objective problem.
• A highly constrained NP-hard problem is solved by an imperialist competitive algorithm.
• The results show that the implemented meta-heuristic outperform the genetic algorithm.

This paper deals with mixed-model assembly line balancing and worker assignment problem (MMALBWAP). Mixed-model assembly lines allow the simultaneous assemble of a set of products on a single assembly line and have achieved great attention during last decades. The worker assignment problem deals with assigning workers to workstations considering their abilities and operating costs. The proposed model in this paper considers two incoherent objectives. The first objective aims to minimize the total cycle time. From one side different models of product have different operating task times and on the other hand different worker skills make more varieties in operating times, therefore minimizing cycle time in these problems seems so important. Simultaneous with cycle time the operating costs related to workers is the second objective of interest to be minimized. In order to solve this multi-objective problem a goal programming approach is utilized and because of high complexity of the problem, an evolutionary algorithm named imperialist competitive algorithm (ICA) is developed. In order to evaluate the efficiency of proposed algorithm, the experimental results obtained are compared with a genetic algorithm.