Parallel evolutionary algorithms for the reconfigurable transfer line balancing problem

This paper deals with an industrial problem of machining line design, which consists in partitioning a given set operations into several subsets corresponding to workstations and sequencing the satisfy technical requirements achieve best performance line. The has complex constraints that include partial order on operations, part positioning, inclusion, exclusion, cycle time, installation parallel machines workstation. is NP-hard even finding feasible solution can be difficult task from practical point view. A evolutionary algorithm (EA) proposed implemented for execution Graphics Processing Unit (GPU). parallelization EA done by working parents one iteration multiple application mutation operator same parent produce offspring. approach evaluated large scale instances demonstrated superior compared algorithms literature terms running time ability obtain solutions. It shown comparison traditional populational scheme newly more suitable advanced GPUs number cores.