Translation of Linear minimization probelm from matlab to r
On Tue, May 08, 2012 at 10:21:59AM -0700, Haio wrote:
Hi everyone, i?m a new user of R and i?m trying to translate an linear
optimization problem from Matlab into r.
The matlab code is as follow:
options = optimset('Diagnostics','on');
[x fval exitflag] = linprog(f,A,b,Aeq,beq,lb,ub,[],options);
exitflag
fval
x=round(x);
Where:
f = Linear objective function vector (vector of 45,rows)
A = Matrix for linear inequality constraints (3colums 45 rows matrix)
b = Vector for linear inequality constraints (3 rows vector)
Aeq = Matrix for linear equality constraints (45 colums, 8 rows )
beq = Vector for linear equality constraints (8 rows vector)
lb =Vector of lower bounds (45 rows)
ub = Vector of upper bounds (45 rows)
I have tryed the package "linprog" although i can?t find anyway to include
the linear inequality constraints into the "solveLP" function.
Hi.
I am not sure, what is the problem with the constraints. According to
the documentation of the function solveLP(), the constraints are
provided as the arguments "bvec" and "Amat". However, i am using "lpSolve"
package and have no experience with "linprog" package. An example
of solving a simple linear program using lpSolve follows.
Consider the problem to maximize 3*x + 2*y with the constraints
x >= 0
y >= 0
x + y <= 3
2*x + y <= 5
x + 2*y <= 5
then try
library(lpSolve)
crit <- c(3, 2)
mat <- rbind(c(1, 1), c(1, 2), c(2, 1))
rhs <- c(3, 5, 5)
dir <- rep("<=", times=3)
out <- lp("max", objective.in=crit, const.mat=mat, const.dir=dir, const.rhs=rhs)
out
Success: the objective function is 8
out$solution
[1] 2 1
Hope this helps.
Petr Savicky.