Error using function. I get the error Array indices must be positive integers or logical values.

clear all
clc
format short
lambda = 0.9;
ES     = zeros(5);
SCV    = ES; %initializing SCV
TH     = ES;  %initializing TH
EW     = ES;  %initializing EW
% a      = 1:5;
% b      = a;
% nodes  = create_nodes(a, b);

for i=1:5
    for j=1:5
        ES(i,j)  = 5-abs(i-3)-abs(j-3); %compute service time
        SCV(i,j) = abs((i-3)*(j-3));    %compute scv
    end
end

EW(1,1) = ES(1,1)*(1+0.5*(1+SCV(1,1))*0.9*1*ES(1,1)/(1-0.9*1*ES(1,1)));
V(1,1)  = 1;
V(1,2)  = 0.5;
EW(1,2) = ES(1,2)*(1+0.5*(1+SCV(1,2))*0.9*V(1,2)*ES(1,2)/(1-0.9*V(1,2)*ES(1,2)));
x(4) = 0.5;
x(7) = 0.5;
x(9) = 0.5;

V_init = 0.5;
for i = 3:5   % compute sojourn time for row 1
    V_fun{i}  = @(x) V_init*x(i-2);
    EW_fun{i} = @(x) ES(1,i)*(1+0.5*(1+SCV(1,i))*0.9*V_fun{i}*ES(1,i)/(1-0.9*V_fun{i}*ES(1,i)));
    V_init    = @(x) V_fun{i};
    fun{i}    = @(x) V_fun{i}*EW_fun{i};
end

V_init = 1;
for j = 6:8   % compute sojourn time for row 2
    V_fun{j}  = @(x) 0.5*(1-x(1))*V_init*x(j-2);
    EW_fun{j} = @(x) ES(2,j-3)*(1+0.5*(1+SCV(2,j-3))*0.9*V_fun{j}*ES(2,j-3)/(1-0.9*V_fun{j}*ES(2,j-3)));
    V_init    = @(x) V_fun(j);
%     fun{j}    = @(x) V_fun(j)*EW_fun{j};
end

V_init = 0.5*x(1)*(1-x(2))*(1-x(5));
for i = 9:10  % compute sojourn time for row 3
    V_fun{i}  = @(x) V_init*x(i-2);
    EW_fun{i} = @(x) ES(3,i-5)*(1+0.5*(1+SCV(1,i-5))*0.9*V_fun{i}*ES(1,i-5)/(1-0.9*V_fun{i}*ES(1,i-5)));
    V_init    = @(x) V_fun{i};
%     fun{i}    = @(x) V_fun{i}*EW_fun{i};
end

V_fun{11}  = @(x) 0.5*x(1)*x(2)*(1-x(3))*(1-x(6))*(1-x(8))*x(9);
EW_fun{11} = @(x) ES(4,5)*(1+0.5*(1+SCV(4,5))*0.9*V_fun{11}*ES(4,5)/(1-0.9*V_fun{11}*ES(4,5)));
% fun{11}    = @(x) V_fun{11}*EW_fun{11};

V_fun{12}  = @(x) 0.5*x(1)*x(2)*x(3)*x(10);
EW_fun{12} = @(x) ES(5,5)*(1+0.5*(1+SCV(5,5))*0.9*V_fun{12}*ES(5,5)/(1-0.9*V_fun{12}*ES(5,5)));
% fun{12}    = @(x) V_fun{12}*EW_fun{12};


% funMin     = @(x) sum([fun{:}])
V_fun{1} = V(1,1);
V_fun{2} = V(1,2);
EW_fun{1} = EW(1,1);
EW_fun{2} = EW(1,2);
% fun{1}    = @(x) V_fun{1}*EW_fun{1};
% fun{2}    = @(x) V_fun{2}*EW_fun{2}
fun       = @(x) cellfun(@(g,h)(g(x)*(h(x))),V_fun,EW_fun);
funMin    = @(x) sum([fun{:}]);

x0   = zeros(1,10);
Aeq  = [];
beq  = [];
lb   = zeros(1,10);
ub   = ones(1,10);
x    = fmincon(fun,x0,[],[],Aeq,beq,lb,ub);

Answer 1

[Walter Roberson]

You have V(1,1) and you assign that to V_fun{1}. We do not know the datatype of V but we do know that indexing with () can never return a function handle, so if V is a variable then V_fun{1} cannot be a function handle. However you use it as if it were a function handle in your g(x) reference in your cellfun.

It is possible but not especially likely that V is a function of two arguments that returns a function handle: that could work, but I doubt it is the situation you have.

More likely, V(1,1) is either a numeric value or else a scalar cell array. Then when you use g(x) that would be a request to index the scalar at location x which is not likely to work.

My guess is that V is a cell array of function handles and that you should be assigning V{1,1} instead of V(1,1)

... Notice how much shorter this reply could have been if you had given us more information.

[Walter Roberson]

 V(1,1)  = 1;
 V(1,2)  = 0.5;

But you try to use those as function handles. You assign the numeric values into the cell and then pull entries from the cell, getting the numeric values, and try to g(x) where g is the numeric value.