CODEWITHSUNDEEP
rlinear-programming
Jose Miranda
Jose Miranda

Reputation: 97

Error in assignment maximizing minimal quality in R language from the book example

I was trying run the example given in the textbook named "Modeling and Solving Linear Programming with R" the link to the textbook can be found here

I tried copying the exact code from the chapter 3.8 on page 75 but it seems that it prints out different result

Assignment01 <- function(c){
n <- dim(c)[1]
coef <- as.vector(t(c))
rhs <- rep(1,2*n)

Amatrix <- matrix(0, 2*n, n*n)

for(i in 1:n){
    for(j in 1:n){
        Amatrix[i, n*(i-1)+j] <- 1
    }
}

for(i in 1:n){
    for(j in 1:n){
        Amatrix[n+i, n*(j-1)+i] <- 1
    }
}

signs <- rep("==", 2*n)
var_type <- rep("B", 2*n)
library(Rglpk)

solution <- Rglpk_solve_LP(obj=coef, mat=Amatrix, dir=signs, types=var_type, rhs=rhs, max=TRUE)
return(solution)

}

Assignment02 <- function(c){
n <- dim(c)[1]
coef <- c(rep(0, n*n),1)
rhs <- c(rep(1, 2*n), rep(0,n))
Amatrix <- matrix(0, 3*n, n*n + 1)

for(i in 1:n){
    for(j in 1:n){
        Amatrix[i, n*(i-1)+j] <- 1
    }
}

for(i in 1:n){
    for(j in 1:n){
        Amatrix[n+i, n*(j-1)+i] <- 1
    }
}

for(i in 1:n){
    for(j in 1:n){
        Amatrix[2*n+1, n*(j-1)+i] <- c[j,i]
    }
}

for(i in 1:n){
    Amatrix[2*n+1, n*n + 1] <- -1

}

signs <- c(rep("==", 2*n),rep(">=",n))

var_type <- c(rep("B", n*n), "C")

library(Rglpk)

solutionPL <- Rglpk_solve_LP(obj=coef, mat=Amatrix, dir=signs, types=var_type, rhs=rhs, max=TRUE)
return (solutionPL) 
}


set.seed(1)
c <- matrix(sample(10:100, 25),100,100)
solAss01 <- Assignment01(c)
m.01 <- matrix(solAss01$solution[1:25],5,5, byrow=TRUE)
solAss02 <- Assignment02(c)
m.02 <- matrix(solAss02$solution[1:25],5,5,byrow=TRUE)
print(m.01)
print(m.02)

and got this output

enter image description here

that should be enter image description here

enter image description here

Upvotes: 0

Views: 66

Answers (1)

Erwin Kalvelagen
Erwin Kalvelagen

Reputation: 16752

This code does not compile:

for(j in 1:){

I believe you are confusing x(i,j) with c(i,j). The x values are zeros and ones with only one 1 in each row and column.

Also notice your c(i,j) have the same value for each j:

> c[1:5,1:5]
     [,1] [,2] [,3] [,4] [,5]
[1,]   34   34   34   34   34
[2,]   43   43   43   43   43
[3,]   60   60   60   60   60
[4,]   89   89   89   89   89
[5,]   27   27   27   27   27

Upvotes: 1

Related Questions