|Institution:||Oregon State University|
|Keywords:||Printed circuits – Design and construction – Mathematical models|
|Full text PDF:||http://hdl.handle.net/1957/30048|
Over the last two decades, the assembly of printed circuit boards (PCB) has generated a huge amount of industrial activity. One of the major developments in PCB assembly was introduction of surface mount technology (SMT). SMT has displaced through-hole technology as a primary means of assembling PCB over the last decade. It has also made it easy to automate PCB assembly process. The component placement machine is probably the most important piece of manufacturing equipment on a surface mount assembly line. It is used for placing components reliably and accurately enough to meet the throughput requirements in a cost-effective manner. Apart from the fact that it is the most expensive equipment on the PCB manufacturing line, it is also often the bottleneck. There are a quite a few areas for improvements on the machine, one of them being component placement sequencing. With the number of components being placed on a PCB ranging in hundreds, a placement sequence which requires near minimum motion of the placement head can help optimize the throughput rates. This research develops an application using genetic algorithm (GA) to solve the component placement sequencing problem for a single headed placement machine. Six different methods were employed. The effects of two parameters which are critical to the execution of a GA were explored at different levels. The results obtained show that the one of the methods performs significantly better than the others. Also, the application developed in this research can be modified in accordance to the problems or machines seen in the industry to optimize the throughput rates.