A simple algorithm, developed for a least-weight structural optimization problem, is used to force the selection of the same
n
n
components of the vectors
X
X
and
Y
Y
, containing
b
b
elements
(
b
>
n
)
(b > n)
so that the objective function
L
~
max
x
i
,
y
i
{
|
X
|
,
|
Y
|
}
\tilde L {\max _{xi,yi}}\left \{ {\left | X \right |,\left | Y \right |} \right \}
is minimized subject to
n
n
equality constraints on each vector,
A
X
=
b
1
AX = {b_1}
,
A
Y
=
b
2
AY = {b_2}
. The method has an obvious advantage over integer programming or branch-and-bound techniques that would, in this case, seek the best selection of
n
n
out of
b
b
elements which satisfy the constraints.