A genetic algorithms enhanced planning system for surface mount PCB assembly

Printed circuit board (PCB) assembly forms the core of a vast array of contemporary manufactured products. The requirements for higher component densities on PCBs, rigid functional specifications, smaller packages and greater reliability, move the electronics manufacturing industry towards automating PCB assembly. Owing to the complexity of PCBs, it becomes impractical to plan the assembly sequence manually. This paper presents a prototype genetic algorithms (GAs) enhanced planning system for surface mount PCB assembly. The prototype system uses GAs to generate the component placement sequence and the component-feeder arrangement for a rotary disk turret, concurrent pick and place SMD machine equipped with a time-delay function. The sequencing process is formulated as a multi-objective optimisation problem under constraints. A framework of the prototype system and the derivation of the multi-objective function are described. The prototype system was validated using examples gleaned from literature. Details of the validation are reported.