A genetic optimization approach to operation of a multi-head surface mounting machine

In this paper, as a practical application, we focus on the genetic algorith m (GA) for multi-head surface mounting machines which are used to populate printed circuit boards (PCBs). Although there have been numerous studies on the surface mounting machine, studies on the multi-head case are rare beca use of its complexity. The multi-head surface mounting machine can pick mul tiple components simultaneously in one pickup operation and this operation can reduce much portion of the assembly time. Hence we try to minimize the assembly time by maximizing the number of simultaneous pickups, resulting i n reduction of PCB production cost. This research introduces a partial-link GA method for the single-head case. Then, we apply this method to the mult i-head case by regarding a reel-group as one reel and a component-cluster a s one component. The results of computer simulation show that our genetic a lgorithm is greatly superior to the heuristic algorithm that is currently u sed in industry.