A very simple question for GA

Dear all, I have a very simple question as follows, I just want to clarify if I am doing right or not:

I have a function named: function [wRemAlo] = Resource allocation (x)

x: have three possible values which I am entering in command line before running the function e.g. I put "x = [90,20,0,10]" and

After receiving the input values of x, my function allocate resources to 16 variables v (v(1) to v(16)) and return the single variable of "wRemAlo".

I want ga to minimise the output of the function: "wRemAlo"

My function is working correct and return the correct value for "wRemAlo" whenever I run it.

I put constraints as follows to put restrictions on ga decisions and pass the handle for the function to ga solver as follows:

ObjectiveFunction = @ResourceAllocation;

nvars = 16; A=[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 ; 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ; 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0 ; 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0 ; 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1 ; 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0 ; 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0 ; 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0 ; 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1 ];

b=[105;90;20;0;10;0;0;45;60]; LB = [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0];

[v,fval] = ga(ObjectiveFunction,nvars,A,b,[],[],LB,[],[])

The fval value returns after running ga dose not make any sense! Should this value be as same as the value for the function output: "wRemAlo"

Please somebody tell me if I am doing correct or not??

regs, Melody

Answer 1

[Seth DeLand]

It looks like there is an issue with the way your decision variables are defined. The input to the objective function (which is currently x), should be a vector of the decision variables. If the v's are your decision variables that you want to optimize, then the objective function should look like

[wRemAlo] = ResourceAllocation(v)

If you also need to use the values in x inside ResourceAllocation, there's an example here that shows how to pass additional arguments (in the minimizing using additional arguments section).

Answer 2

[Nagh Meh]

Thanks a lot Seth,

I changed it to followings:

==================================== w1 = 0; w2 = 0; w3 = 45; w4 = 60;

x1 = 90; x2 = 20; x3 = 0; x4 = 10;

ObjectiveFunction = @(v) ResourceAllocation (v,x1,x2,x3,x4,w1,w2,w3,w4);

nvars = 16; A=[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 ;

   1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ; 
   0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0 ; 
   0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0 ; 
   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1 ;

   1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0 ;
   0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0 ;
   0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0 ;
   0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1 ];

b=[w1+w2+w3+w4;x1;x2;x3;x4;w1;w2;w3;w4]; LB = [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0];

[v,fval] = ga(ObjectiveFunction,nvars,A,b,[],[],LB,[],[]) ====================================

I also need to clarify one thing:

In my fitness function:

[wRemAlo] = ResourceAllocation(v,x1,x2,x3,x4,w1,w2,w3,w4)

v(1), v(2), ..... v(16) are my decision variables that I want ga to find the best values for them with which "wRemAlo" would be minimised.

In my fitness function, I calculate the values for v through loops and fors and I assigned the calcuted value at the end of the function to v(1) to v(16) e.g. :

v(1) = H1_all_w1; v(2) = H1_all_w2; v(3) = H1_all_w3; . . . . .

With which I have mentioned how the values for all v() s will be calculated through the function.

and when I call the function from the CL:

[v] = ResourceAllocation(v,x1,x2,x3,x4,w1,w2,w3,w4)

the values for vs are:

Columns 1 through 9

     0     0    15    20     0     0    15     5     0

Columns 10 through 16

     0     0     0     0     0    10     0

I would like to know, am I doing correct or I do not need to assign the values for vs inside the pure function as it may give me the wrong answers?

regs, M.