Component allocation in printed circuit card assembly systems is a spe
cial case of the classical mixed-model assembly line balancing problem
and involves assigning component types to machines to achieve specifi
c production objectives. In this paper the component allocation proble
m is considered for the scenario where there are two or more placement
machines (possibly nonidentical) and the objective is to balance, for
every card type, a combination of the card assembly time and the mach
ine setup time. A mathematical formulation of the problem is developed
for a class of placement machines. Two alternative solution approache
s are presented: a list-processing-based heuristic for a simple versio
n of the problem, and a linear-programming-based branch-and-bound proc
edure for the general component allocation problem. Industrial case st
udy results are presented for each approach that indicate expected thr
oughput improvements of up to 8-10% over the company's current procedu
re, with much less direct effort required by the process engineer.