GA - Quadratic Assignment Problem

Hey guys,

I am using GA to solve an optimization problem, i.e. quadratic assignment problem (QAP).

the objective function of QAP is

min EEEE abxx + cx where E is summation A, B , C is input matrix and is constant

the only variable here is X, which is a permutation matrix (a binary matrix with the sum of each row and each column is equal to1)

i have tested my objective function in a script, and it works. and i can use randperm and a few lines of coding to generate permutation matrix

the problem is, when i implement GA with optimization toolbox, it always output an error - "too many input arguments"

does anyone know what causes this to be happened?

and should i put my variables(permutation matrix) at constraint pane or creation function pane?

how matlab know the feasible solution can only be a permutation matrix, so that after crossover and mutation, the children will still be a permutation matrix?

does any guys have any idea to implement this QAP?

thanks in advanced.

[Wee Loon]

can't anybody please help me?

[Andrew Newell]

@Wee, it's hard to know what the problem is unless you show us what code you're using to do this calculation.