An Efficient Algorithm for Placement Sequence and Feeder Assignment Problems With Multiple Placement-Nozzles and Independent Link Evaluation

We show that the genetic algorithm that Leu described to plan component placement sequences and feeder assignments for pick-and-place printed circuit board assembly tasks is a special case of a more general model that supports multiple placement-nozzles and independent feeder and board link (chromosome) evaluation methods. We also show that independent link evaluation can be used to offset a reduction in the parent link sample space and that these results are better than what can be achieved through link-pair evaluation. These generalizations extend the capabilities of the genetic algorithm to a broader range of manufacturing scenarios.